WO2024058488A1 - System for providing real-time sleep health management service by using ai-based brain wave synchronization and autonomic nervous system control - Google Patents
System for providing real-time sleep health management service by using ai-based brain wave synchronization and autonomic nervous system control Download PDFInfo
- Publication number
- WO2024058488A1 WO2024058488A1 PCT/KR2023/013327 KR2023013327W WO2024058488A1 WO 2024058488 A1 WO2024058488 A1 WO 2024058488A1 KR 2023013327 W KR2023013327 W KR 2023013327W WO 2024058488 A1 WO2024058488 A1 WO 2024058488A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- sleep
- time
- user
- wake
- brain wave
- Prior art date
Links
- 230000007958 sleep Effects 0.000 title claims abstract description 388
- 210000004556 brain Anatomy 0.000 title claims abstract description 178
- 210000003403 autonomic nervous system Anatomy 0.000 title claims abstract description 87
- 230000036541 health Effects 0.000 title claims abstract description 65
- 238000000034 method Methods 0.000 claims abstract description 95
- 230000029058 respiratory gaseous exchange Effects 0.000 claims abstract description 59
- 210000001186 vagus nerve Anatomy 0.000 claims abstract description 26
- YJPIGAIKUZMOQA-UHFFFAOYSA-N Melatonin Natural products COC1=CC=C2N(C(C)=O)C=C(CCN)C2=C1 YJPIGAIKUZMOQA-UHFFFAOYSA-N 0.000 claims abstract description 19
- 229960003987 melatonin Drugs 0.000 claims abstract description 19
- DRLFMBDRBRZALE-UHFFFAOYSA-N melatonin Chemical compound COC1=CC=C2NC=C(CCNC(C)=O)C2=C1 DRLFMBDRBRZALE-UHFFFAOYSA-N 0.000 claims abstract description 19
- 230000000638 stimulation Effects 0.000 claims abstract description 17
- 230000007383 nerve stimulation Effects 0.000 claims abstract description 15
- 210000001002 parasympathetic nervous system Anatomy 0.000 claims abstract description 12
- 230000006698 induction Effects 0.000 claims abstract description 10
- 230000003238 somatosensory effect Effects 0.000 claims abstract description 8
- 238000004458 analytical method Methods 0.000 claims description 184
- 230000008667 sleep stage Effects 0.000 claims description 142
- 238000007726 management method Methods 0.000 claims description 90
- 230000004622 sleep time Effects 0.000 claims description 70
- 238000013473 artificial intelligence Methods 0.000 claims description 63
- 230000002618 waking effect Effects 0.000 claims description 56
- 230000036385 rapid eye movement (rem) sleep Effects 0.000 claims description 43
- 230000000903 blocking effect Effects 0.000 claims description 40
- 206010062519 Poor quality sleep Diseases 0.000 claims description 27
- 230000027288 circadian rhythm Effects 0.000 claims description 24
- 208000013738 Sleep Initiation and Maintenance disease Diseases 0.000 claims description 21
- 206010022437 insomnia Diseases 0.000 claims description 21
- 208000001491 myopia Diseases 0.000 claims description 14
- 230000004379 myopia Effects 0.000 claims description 14
- 201000009310 astigmatism Diseases 0.000 claims description 13
- 230000007423 decrease Effects 0.000 claims description 13
- 238000004422 calculation algorithm Methods 0.000 claims description 12
- 208000022249 Sleep-Wake Transition disease Diseases 0.000 claims description 11
- 208000007590 Disorders of Excessive Somnolence Diseases 0.000 claims description 10
- 230000006399 behavior Effects 0.000 claims description 10
- 206010020765 hypersomnia Diseases 0.000 claims description 10
- 230000033228 biological regulation Effects 0.000 claims description 8
- 208000010340 Sleep Deprivation Diseases 0.000 claims description 7
- 208000017164 Chronobiology disease Diseases 0.000 claims description 6
- 210000000707 wrist Anatomy 0.000 claims description 6
- 208000013403 hyperactivity Diseases 0.000 claims description 3
- 230000002401 inhibitory effect Effects 0.000 claims description 3
- 210000002820 sympathetic nervous system Anatomy 0.000 abstract description 2
- 230000033001 locomotion Effects 0.000 description 45
- 230000006870 function Effects 0.000 description 41
- 230000000694 effects Effects 0.000 description 29
- 238000013136 deep learning model Methods 0.000 description 27
- 238000010586 diagram Methods 0.000 description 24
- 238000004364 calculation method Methods 0.000 description 21
- 238000013528 artificial neural network Methods 0.000 description 20
- 230000035882 stress Effects 0.000 description 20
- 238000004891 communication Methods 0.000 description 18
- 238000000537 electroencephalography Methods 0.000 description 18
- QZAYGJVTTNCVMB-UHFFFAOYSA-N serotonin Chemical compound C1=C(O)C=C2C(CCN)=CNC2=C1 QZAYGJVTTNCVMB-UHFFFAOYSA-N 0.000 description 16
- 238000013135 deep learning Methods 0.000 description 15
- 238000010801 machine learning Methods 0.000 description 15
- 230000008569 process Effects 0.000 description 15
- 238000001126 phototherapy Methods 0.000 description 14
- 208000019901 Anxiety disease Diseases 0.000 description 12
- 208000006561 Cluster Headache Diseases 0.000 description 12
- 208000002193 Pain Diseases 0.000 description 12
- 230000008859 change Effects 0.000 description 11
- 238000012549 training Methods 0.000 description 11
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 10
- 206010016256 fatigue Diseases 0.000 description 10
- 238000012790 confirmation Methods 0.000 description 9
- 201000010099 disease Diseases 0.000 description 9
- 238000012545 processing Methods 0.000 description 9
- 230000009471 action Effects 0.000 description 8
- 230000036506 anxiety Effects 0.000 description 8
- 208000010877 cognitive disease Diseases 0.000 description 8
- 230000001965 increasing effect Effects 0.000 description 8
- 238000012544 monitoring process Methods 0.000 description 8
- 229940076279 serotonin Drugs 0.000 description 8
- 230000036772 blood pressure Effects 0.000 description 7
- 238000001514 detection method Methods 0.000 description 7
- 238000009826 distribution Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 7
- 210000001652 frontal lobe Anatomy 0.000 description 7
- 230000001939 inductive effect Effects 0.000 description 7
- 230000036407 pain Effects 0.000 description 7
- 230000033764 rhythmic process Effects 0.000 description 7
- 238000001228 spectrum Methods 0.000 description 7
- 208000019116 sleep disease Diseases 0.000 description 6
- 230000003860 sleep quality Effects 0.000 description 6
- 208000024891 symptom Diseases 0.000 description 6
- 208000006096 Attention Deficit Disorder with Hyperactivity Diseases 0.000 description 5
- 208000036864 Attention deficit/hyperactivity disease Diseases 0.000 description 5
- 208000020925 Bipolar disease Diseases 0.000 description 5
- 208000000094 Chronic Pain Diseases 0.000 description 5
- 210000004027 cell Anatomy 0.000 description 5
- 230000001276 controlling effect Effects 0.000 description 5
- 238000013527 convolutional neural network Methods 0.000 description 5
- 238000013480 data collection Methods 0.000 description 5
- 229940088597 hormone Drugs 0.000 description 5
- 239000005556 hormone Substances 0.000 description 5
- 230000001976 improved effect Effects 0.000 description 5
- 208000027061 mild cognitive impairment Diseases 0.000 description 5
- 210000005036 nerve Anatomy 0.000 description 5
- 238000007781 pre-processing Methods 0.000 description 5
- 230000028327 secretion Effects 0.000 description 5
- 230000036578 sleeping time Effects 0.000 description 5
- 208000024827 Alzheimer disease Diseases 0.000 description 4
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 4
- 230000001133 acceleration Effects 0.000 description 4
- 208000018912 cluster headache syndrome Diseases 0.000 description 4
- 230000003203 everyday effect Effects 0.000 description 4
- 210000003016 hypothalamus Anatomy 0.000 description 4
- 210000002569 neuron Anatomy 0.000 description 4
- 210000001747 pupil Anatomy 0.000 description 4
- 238000011084 recovery Methods 0.000 description 4
- 230000000306 recurrent effect Effects 0.000 description 4
- 201000000980 schizophrenia Diseases 0.000 description 4
- 230000001953 sensory effect Effects 0.000 description 4
- 238000003860 storage Methods 0.000 description 4
- 230000001360 synchronised effect Effects 0.000 description 4
- 208000028698 Cognitive impairment Diseases 0.000 description 3
- 206010019233 Headaches Diseases 0.000 description 3
- 208000018737 Parkinson disease Diseases 0.000 description 3
- 238000013459 approach Methods 0.000 description 3
- 230000002146 bilateral effect Effects 0.000 description 3
- 210000004204 blood vessel Anatomy 0.000 description 3
- 230000001413 cellular effect Effects 0.000 description 3
- 230000002354 daily effect Effects 0.000 description 3
- 229940079593 drug Drugs 0.000 description 3
- 239000003814 drug Substances 0.000 description 3
- 230000000517 effect on sleep Effects 0.000 description 3
- 230000004424 eye movement Effects 0.000 description 3
- 231100000869 headache Toxicity 0.000 description 3
- 230000001788 irregular Effects 0.000 description 3
- 230000007774 longterm Effects 0.000 description 3
- 230000015654 memory Effects 0.000 description 3
- 230000001537 neural effect Effects 0.000 description 3
- 230000008452 non REM sleep Effects 0.000 description 3
- 239000000902 placebo Substances 0.000 description 3
- 229940068196 placebo Drugs 0.000 description 3
- 208000028173 post-traumatic stress disease Diseases 0.000 description 3
- 230000004044 response Effects 0.000 description 3
- 230000003938 response to stress Effects 0.000 description 3
- 239000013589 supplement Substances 0.000 description 3
- 230000002889 sympathetic effect Effects 0.000 description 3
- 238000002560 therapeutic procedure Methods 0.000 description 3
- 210000001835 viscera Anatomy 0.000 description 3
- 206010002091 Anaesthesia Diseases 0.000 description 2
- 241000282412 Homo Species 0.000 description 2
- 208000019022 Mood disease Diseases 0.000 description 2
- 208000021384 Obsessive-Compulsive disease Diseases 0.000 description 2
- 230000001154 acute effect Effects 0.000 description 2
- VREFGVBLTWBCJP-UHFFFAOYSA-N alprazolam Chemical compound C12=CC(Cl)=CC=C2N2C(C)=NN=C2CN=C1C1=CC=CC=C1 VREFGVBLTWBCJP-UHFFFAOYSA-N 0.000 description 2
- 230000037005 anaesthesia Effects 0.000 description 2
- 230000003416 augmentation Effects 0.000 description 2
- 210000000467 autonomic pathway Anatomy 0.000 description 2
- 210000000988 bone and bone Anatomy 0.000 description 2
- 210000000621 bronchi Anatomy 0.000 description 2
- 229910002092 carbon dioxide Inorganic materials 0.000 description 2
- 239000001569 carbon dioxide Substances 0.000 description 2
- 230000002060 circadian Effects 0.000 description 2
- 239000003086 colorant Substances 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 230000003247 decreasing effect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 208000002173 dizziness Diseases 0.000 description 2
- 230000008482 dysregulation Effects 0.000 description 2
- 230000002996 emotional effect Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000005176 gastrointestinal motility Effects 0.000 description 2
- JYGXADMDTFJGBT-VWUMJDOOSA-N hydrocortisone Chemical compound O=C1CC[C@]2(C)[C@H]3[C@@H](O)C[C@](C)([C@@](CC4)(O)C(=O)CO)[C@@H]4[C@@H]3CCC2=C1 JYGXADMDTFJGBT-VWUMJDOOSA-N 0.000 description 2
- 238000005304 joining Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000002156 mixing Methods 0.000 description 2
- 238000010295 mobile communication Methods 0.000 description 2
- 210000003205 muscle Anatomy 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 230000010355 oscillation Effects 0.000 description 2
- 208000019906 panic disease Diseases 0.000 description 2
- 210000005037 parasympathetic nerve Anatomy 0.000 description 2
- 230000000144 pharmacologic effect Effects 0.000 description 2
- 238000000718 qrs complex Methods 0.000 description 2
- 230000003252 repetitive effect Effects 0.000 description 2
- 230000037322 slow-wave sleep Effects 0.000 description 2
- 210000000278 spinal cord Anatomy 0.000 description 2
- 230000000153 supplemental effect Effects 0.000 description 2
- 210000004243 sweat Anatomy 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 210000000115 thoracic cavity Anatomy 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 101150103244 ACT1 gene Proteins 0.000 description 1
- 206010003694 Atrophy Diseases 0.000 description 1
- 240000007124 Brassica oleracea Species 0.000 description 1
- 235000003899 Brassica oleracea var acephala Nutrition 0.000 description 1
- 235000012905 Brassica oleracea var viridis Nutrition 0.000 description 1
- 102100031102 C-C motif chemokine 4 Human genes 0.000 description 1
- 101710155855 C-C motif chemokine 4 Proteins 0.000 description 1
- VYZAMTAEIAYCRO-UHFFFAOYSA-N Chromium Chemical compound [Cr] VYZAMTAEIAYCRO-UHFFFAOYSA-N 0.000 description 1
- 208000019888 Circadian rhythm sleep disease Diseases 0.000 description 1
- 206010057315 Daydreaming Diseases 0.000 description 1
- 206010012289 Dementia Diseases 0.000 description 1
- 101710140859 E3 ubiquitin ligase TRAF3IP2 Proteins 0.000 description 1
- 102100026620 E3 ubiquitin ligase TRAF3IP2 Human genes 0.000 description 1
- 208000027534 Emotional disease Diseases 0.000 description 1
- 206010016754 Flashback Diseases 0.000 description 1
- PEDCQBHIVMGVHV-UHFFFAOYSA-N Glycerine Chemical compound OCC(O)CO PEDCQBHIVMGVHV-UHFFFAOYSA-N 0.000 description 1
- 101100161918 Glycine max SAC1 gene Proteins 0.000 description 1
- 206010020772 Hypertension Diseases 0.000 description 1
- 206010048533 Hypervigilance Diseases 0.000 description 1
- 208000001456 Jet Lag Syndrome Diseases 0.000 description 1
- 206010024264 Lethargy Diseases 0.000 description 1
- 208000016285 Movement disease Diseases 0.000 description 1
- 241000699670 Mus sp. Species 0.000 description 1
- 102000007399 Nuclear hormone receptor Human genes 0.000 description 1
- 108020005497 Nuclear hormone receptor Proteins 0.000 description 1
- 208000025174 PANDAS Diseases 0.000 description 1
- 208000021155 Paediatric autoimmune neuropsychiatric disorders associated with streptococcal infection Diseases 0.000 description 1
- 240000000220 Panda oleosa Species 0.000 description 1
- 235000016496 Panda oleosa Nutrition 0.000 description 1
- 208000012545 Psychophysiologic disease Diseases 0.000 description 1
- 238000012952 Resampling Methods 0.000 description 1
- 241000283984 Rodentia Species 0.000 description 1
- 206010067493 Sleep inertia Diseases 0.000 description 1
- 206010041349 Somnolence Diseases 0.000 description 1
- 241000863032 Trieres Species 0.000 description 1
- 230000003213 activating effect Effects 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 230000003321 amplification Effects 0.000 description 1
- 210000004727 amygdala Anatomy 0.000 description 1
- 239000005557 antagonist Substances 0.000 description 1
- 230000037007 arousal Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 1
- 230000037444 atrophy Effects 0.000 description 1
- 230000003190 augmentative effect Effects 0.000 description 1
- 230000035559 beat frequency Effects 0.000 description 1
- 230000006736 behavioral deficit Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000002457 bidirectional effect Effects 0.000 description 1
- 239000000090 biomarker Substances 0.000 description 1
- 208000028683 bipolar I disease Diseases 0.000 description 1
- 230000004397 blinking Effects 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 230000036770 blood supply Effects 0.000 description 1
- 230000007177 brain activity Effects 0.000 description 1
- 230000001680 brushing effect Effects 0.000 description 1
- 230000020411 cell activation Effects 0.000 description 1
- 239000003795 chemical substances by application Substances 0.000 description 1
- 210000000038 chest Anatomy 0.000 description 1
- 230000001684 chronic effect Effects 0.000 description 1
- 230000037326 chronic stress Effects 0.000 description 1
- 230000008632 circadian clock Effects 0.000 description 1
- 230000003920 cognitive function Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 230000001054 cortical effect Effects 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 208000035475 disorder Diseases 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 238000004836 empirical method Methods 0.000 description 1
- 230000000763 evoking effect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000007667 floating Methods 0.000 description 1
- 239000000122 growth hormone Substances 0.000 description 1
- 210000003128 head Anatomy 0.000 description 1
- 229960000890 hydrocortisone Drugs 0.000 description 1
- 230000037417 hyperactivation Effects 0.000 description 1
- 230000000147 hypnotic effect Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 208000033915 jet lag type circadian rhythm sleep disease Diseases 0.000 description 1
- 230000009191 jumping Effects 0.000 description 1
- 230000002197 limbic effect Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 210000001259 mesencephalon Anatomy 0.000 description 1
- 230000003278 mimic effect Effects 0.000 description 1
- 230000036651 mood Effects 0.000 description 1
- 210000000653 nervous system Anatomy 0.000 description 1
- 230000007996 neuronal plasticity Effects 0.000 description 1
- 108020004017 nuclear receptors Proteins 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 230000000474 nursing effect Effects 0.000 description 1
- 210000000956 olfactory bulb Anatomy 0.000 description 1
- 230000003534 oscillatory effect Effects 0.000 description 1
- 230000001151 other effect Effects 0.000 description 1
- 229910052760 oxygen Inorganic materials 0.000 description 1
- 239000001301 oxygen Substances 0.000 description 1
- 230000001734 parasympathetic effect Effects 0.000 description 1
- 210000001152 parietal lobe Anatomy 0.000 description 1
- 230000008506 pathogenesis Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000037081 physical activity Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000007639 printing Methods 0.000 description 1
- 238000004393 prognosis Methods 0.000 description 1
- 108090000623 proteins and genes Proteins 0.000 description 1
- 102000004169 proteins and genes Human genes 0.000 description 1
- 238000013138 pruning Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 230000009257 reactivity Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 230000036387 respiratory rate Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000035807 sensation Effects 0.000 description 1
- 229960002078 sevoflurane Drugs 0.000 description 1
- DFEYYRMXOJXZRJ-UHFFFAOYSA-N sevoflurane Chemical compound FCOC(C(F)(F)F)C(F)(F)F DFEYYRMXOJXZRJ-UHFFFAOYSA-N 0.000 description 1
- 208000020685 sleep-wake disease Diseases 0.000 description 1
- 230000004037 social stress Effects 0.000 description 1
- 238000011301 standard therapy Methods 0.000 description 1
- 210000001562 sternum Anatomy 0.000 description 1
- 230000004936 stimulating effect Effects 0.000 description 1
- 230000002739 subcortical effect Effects 0.000 description 1
- 210000000221 suprachiasmatic nucleus Anatomy 0.000 description 1
- 230000008700 sympathetic activation Effects 0.000 description 1
- 210000000225 synapse Anatomy 0.000 description 1
- 208000011580 syndromic disease Diseases 0.000 description 1
- 210000003478 temporal lobe Anatomy 0.000 description 1
- 230000001225 therapeutic effect Effects 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
- 230000010356 wave oscillation Effects 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61M—DEVICES FOR INTRODUCING MEDIA INTO, OR ONTO, THE BODY; DEVICES FOR TRANSDUCING BODY MEDIA OR FOR TAKING MEDIA FROM THE BODY; DEVICES FOR PRODUCING OR ENDING SLEEP OR STUPOR
- A61M21/00—Other devices or methods to cause a change in the state of consciousness; Devices for producing or ending sleep by mechanical, optical, or acoustical means, e.g. for hypnosis
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61M—DEVICES FOR INTRODUCING MEDIA INTO, OR ONTO, THE BODY; DEVICES FOR TRANSDUCING BODY MEDIA OR FOR TAKING MEDIA FROM THE BODY; DEVICES FOR PRODUCING OR ENDING SLEEP OR STUPOR
- A61M21/00—Other devices or methods to cause a change in the state of consciousness; Devices for producing or ending sleep by mechanical, optical, or acoustical means, e.g. for hypnosis
- A61M21/02—Other devices or methods to cause a change in the state of consciousness; Devices for producing or ending sleep by mechanical, optical, or acoustical means, e.g. for hypnosis for inducing sleep or relaxation, e.g. by direct nerve stimulation, hypnosis, analgesia
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61N—ELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
- A61N5/00—Radiation therapy
- A61N5/06—Radiation therapy using light
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
- G16H20/70—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to mental therapies, e.g. psychological therapy or autogenous training
Definitions
- the present invention relates to a real-time sleep health management service provision system using AI-based brain wave entrainment and autonomic nervous system regulation, which normalizes sleep and circadian rhythm by performing brain wave entrainment and autonomic nervous system regulation through light, sound, breathing, and vibration. Provides a system.
- Sleep disorders such as insomnia occur due to various stresses as society develops.
- the Korean Standard Classification of Diseases also classifies sleep disorders with disease code G47.
- Subcategories of G47 include insomnia-related diseases (G47.0) and sleep-wake disorders (G47.2).
- Sufficient sleep not only restores a person's health and vitality, but is also closely related to various hormones in the human body. Therefore, if you do not get adequate sleep, you may feel fatigue and lethargy in daily life, but in severe cases, you may be exposed to serious diseases such as chronic fatigue, dementia, depression, and panic disorder. Therefore, as the rate of sleep disorders among modern people increases, the need for sleep care also increases.
- Brain waves and autonomic nervous system are regulated depending on sleeping and waking up, but in the case of insomnia or excessive sleep, brain waves and autonomic nervous system regulation tend to deviate from the normal range.
- hypothalamus which is responsible for the circadian rhythm
- circadian rhythm not only the circadian rhythm is broken, but it can also cause cluster headaches and cause various symptoms such as depression, ADHD, anxiety disorder, obsessive-compulsive disorder, and panic disorder. This can be achieved. Accordingly, research and development of a system that can ultimately restore circadian rhythm by managing the sleep cycle is required.
- One embodiment of the present invention enables brain wave synchronization using light and sound, and regulates the autonomic nervous system using light and breathing, thereby adjusting the brain waves and autonomic nervous system to match bedtime and wake-up time.
- a monaural beat or binaural beat sound source that synchronizes theta waves, alpha waves, and beta waves is played through the speaker to induce brain waves when waking up, and a blue light sunlight shower is provided at the waking time to wake up.
- a blue light sunlight shower is provided at the waking time to wake up.
- it generates serotonin, and the generated serotonin is converted into melatonin at bedtime, which can further induce sleep and sound sleep.
- a shift agent is used to lower the stress.
- Sensory stimulation Bi-Lateral Alternating Stimulation in Tactile
- cluster headaches caused by abnormal activity of the hypothalamus are alleviated by normalizing the circadian rhythm
- non-invasive vagus nerve stimulation NonInvasive Vagus
- sleep biometric data for each user collected while the user is sleeping without moving is accumulated through a wearable device such as a smart watch, and then sleep patterns for each user are determined through the accumulated sleep biometric data.
- the purpose is to provide a system and method for providing real-time sleep health management services using AI-based brain wave entrainment and autonomic nervous system control that can analyze and identify optimal sleep cycle information for each user.
- the user's sleep quality is deteriorated by providing a wake-up alarm at a set time without considering the sleep stage or sleep cycle by simply setting the wake-up time, which has a problem with the conventional alarm that wakes the user from sleep.
- a real-time sleep health management service using AI-based brain wave entrainment and autonomic nervous system regulation that can effectively solve problems such as feeling more tired due to waking up time that does not take into account the sleep cycle even after sleeping for a sufficient amount of time. The purpose is to provide systems and methods.
- the optimal sleep cycle information for each user is a generalization of which REM sleep period the user feels most refreshed upon waking up.
- the optimal sleep cycle information for each user provided in the embodiment is biometric information collected according to the waking time. Through analysis, each user can be identified.
- wearable devices such as smart watches are used to classify the user's movement, accumulate heart rate data when there is no movement, and predict the user's sleep stage through deep learning of the accumulated data. Afterwards, the REM sleep stage is determined considering the sleep stage and heart rate change, and a wake-up alarm is provided through the smart watch in the light sleep stage after REM sleep. Additionally, in an embodiment, the alarm may be repeated until the user is confirmed to be awake through the user's movements and heart rate.
- the optimal wake-up time is readjusted to suit the repeated sleep cycle of 1.5 hours to wake the user.
- bedtime and wake-up time that optimize sleep efficiency which is the actual sleeping time compared to the time lying in bed, are calculated through user-specific sleep biometric data learned through machine learning, and these are converted into bedtime alarm and wake-up alarm times. You can set it.
- machine learning based on a deep learning neural network is performed based on a customized training data set optimized for determining sleep stage and identifying optimal cycle information (i.e., optimal sleep cycle information).
- optimal cycle information i.e., optimal sleep cycle information
- one embodiment of the present invention sets the bedtime and wake-up time, outputs a sleep induction method, which is a breathing method that stimulates the parasympathetic nervous system at bedtime, and uses light of a melatonin-inhibiting wavelength.
- a sleep induction method which is a breathing method that stimulates the parasympathetic nervous system at bedtime
- uses light of a melatonin-inhibiting wavelength In order to synchronize brain waves with alpha or beta waves at the time of waking up, an alpha-band or beta-band flicker is inserted and output, and blue light is output to promote awakening.
- a database unit that stores settings for light, sound, vibration and breathing methods for brain wave entrainment and autonomic nervous system control at the user terminal and bedtime and wake-up time, a settings unit that receives settings for bedtime and wake-up time from the user terminal, and the user It includes a management service providing server including a control unit that outputs at least one of light, sound, vibration, and breathing for brain wave entrainment and autonomic nervous system control at the bedtime and wake-up time set in the terminal.
- sleep biometric data for each user collected while the user is sleeping without moving is accumulated through a wearable device such as a smart watch, and then sleep patterns for each user are determined through the accumulated sleep biometric data.
- the purpose is to provide a system and method for providing real-time sleep health management services using AI-based brain wave entrainment and autonomic nervous system control that can analyze and identify optimal sleep cycle information for each user.
- brain wave synchronization is achieved using light and sound
- the autonomic nervous system is controlled using light and breathing, so that brain waves are adjusted to match bedtime and waking time.
- and adjusts the autonomic nervous system induces brain waves when waking up by playing a sound source of monaural beats or binaural beats that synchronize theta waves, alpha waves, and beta waves through a speaker according to the waking time, and uses blue light at the waking time.
- a sunlight shower it generates serotonin along with an awakening effect, and the generated serotonin is converted into melatonin at bedtime to further induce sleep and sound sleep.
- Alternating Somatosensory Stimulation (Bi-Lateral Alternating Stimulation in Tactile) is provided through user terminals and wearable devices, and it relieves cluster headaches caused by abnormal activity of the hypothalamus by normalizing circadian rhythm and is non-invasive. Pain can be alleviated through vagus nerve stimulation (NonInvasive Vagus Nerve Stimulation).
- the system and method for providing a real-time sleep health management service using AI-based brain wave tuning and autonomic nervous system control of the present invention adjusts the user's sleep according to the individual's sleep cycle such as sleep time, bedtime, wake-up time, and nap time.
- the sleep cycle can be used to ensure that the user can sleep refreshed even if he or she sleeps briefly, thereby improving the quality of life by increasing the user's time efficiency, sleep efficiency, and rest efficiency.
- system and method for providing real-time sleep health management services using AI-based brain wave entrainment and autonomic nervous system control of the present invention informs the user of the required REM sleep time if the user does not recover sufficiently after sleep, or By providing various services for user recovery, users can speed up their recovery from stress and fatigue.
- system and method for providing real-time sleep health management services using AI-based brain wave entrainment and autonomic nervous system control of the present invention are based on a deep learning neural network based on a customized training data set optimized for sleep stage determination.
- a sleep stage judgment model By implementing a sleep stage judgment model by performing machine learning, the quality of the sleep stage prediction and wake-up time calculation results calculated from the sleep stage judgment model can be further improved.
- the effects of the present invention are not limited to the above effects, and should be understood to include all effects that can be inferred from the configuration of the invention described in the detailed description or claims of the present invention.
- the effects that can be obtained from the present invention are not limited to the effects described above, and other effects may exist.
- Figure 1 is a diagram for explaining a real-time sleep health management service providing system using AI-based brain wave entrainment and autonomic nervous system control according to the first embodiment of the present invention.
- FIG. 2 is a block diagram illustrating a management service providing server included in the system of FIG. 1.
- Figures 3 and 4 are diagrams for explaining an embodiment in which a real-time sleep health management service using AI-based brain wave entrainment and autonomic nervous system control according to the first embodiment of the present invention is implemented.
- Figure 5 is an operation flowchart illustrating a method of providing a real-time sleep health management service using AI-based brain wave entrainment and autonomic nervous system control according to the first embodiment of the present invention.
- Figure 6 is a diagram showing a typical sleep cycle.
- Figure 7 is a diagram showing the configuration of a system for providing real-time sleep health management service using AI-based brain wave entrainment and autonomic nervous system control according to the second embodiment of the present invention.
- Figure 8 is a diagram showing the data processing configuration of a smart watch according to a second embodiment of the present invention.
- Figure 9 is a diagram showing the data processing configuration of the analysis server according to the second embodiment of the present invention.
- 10 to 14 are diagrams for explaining a deep learning model used in an analysis server according to a second embodiment of the present invention.
- Figure 15 is a signal flow diagram of a real-time sleep health management service providing system using AI-based brain wave entrainment and autonomic nervous system control according to the second embodiment of the present invention.
- Figure 16 is a diagram showing a method of providing a wake-up alarm according to a second embodiment of the present invention.
- Figure 17 is a diagram showing a method of adjusting the wake-up time reflecting the user's actual sleep time according to the second embodiment of the present invention.
- 'part' includes a unit realized by hardware, a unit realized by software, and a unit realized using both. Additionally, one unit may be realized using two or more pieces of hardware, and two or more units may be realized using one piece of hardware.
- ' ⁇ part' is not limited to software or hardware, and ' ⁇ part' may be configured to reside in an addressable storage medium or may be configured to reproduce one or more processors. Therefore, as an example, ' ⁇ part' refers to components such as software components, object-oriented software components, class components, and task components, processes, functions, properties, and procedures. , subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays, and variables.
- components and 'parts' may be combined into a smaller number of components and 'parts' or may be further separated into additional components and 'parts'. Additionally, components and 'parts' may be implemented to regenerate one or more CPUs within a device or a secure multimedia card.
- some of the operations or functions described as being performed by a terminal, apparatus, or device may instead be performed on a server connected to the terminal, apparatus, or device.
- some of the operations or functions described as being performed by the server may also be performed in a terminal, apparatus, or device connected to the server.
- mapping or matching with the terminal mean mapping or matching the terminal's unique number or personal identification information, which is identifying data of the terminal. It can be interpreted as
- FIG. 1 is a diagram for explaining a real-time sleep health management service providing system using AI-based brain wave entrainment and autonomic nervous system control according to the first embodiment of the present invention.
- a real-time sleep health management service providing system 1 using AI-based brain wave tuning and autonomic nervous system control includes at least one user terminal 100, a management service providing server 300, and at least one wearable device. It may include (400).
- the real-time sleep health management service providing system (1) using AI-based brain wave tuning and autonomic nervous system control shown in FIG. 1 is only an embodiment of the present invention, and the present invention is not limited to FIG. 1. .
- each component of FIG. 1 is generally connected through a network (Network, 200).
- Network Network, 200
- at least one user terminal 100 may be connected to the management service providing server 300 through the network 200.
- the management service providing server 300 may be connected to at least one user terminal 100 and at least one wearable device 400 through the network 200.
- at least one wearable device 400 may be connected to the management service providing server 300 through the network 200.
- the network refers to a connection structure that allows information exchange between each node, such as a plurality of terminals and servers.
- Examples of such networks include a local area network (LAN) and a wide area network (WAN). Wide Area Network, Internet (WWW: World Wide Web), wired and wireless data communication network, telephone network, wired and wireless television communication network, etc.
- Examples of wireless data communication networks include 3G, 4G, 5G, 3rd Generation Partnership Project (3GPP), 5th Generation Partnership Project (5GPP), 5G New Radio (NR), 6th Generation of Cellular Networks (6G), and Long Term Evolution (LTE).
- 3GPP 3rd Generation Partnership Project
- 5GPP 5th Generation Partnership Project
- NR 5G New Radio
- 6G 6th Generation of Cellular Networks
- LTE Long Term Evolution
- WIMAX World Interoperability for Microwave Access
- Wi-Fi Internet
- LAN Local Area Network
- Wireless LAN Wireless Local Area Network
- WAN Wide Area Network
- PAN Personal Area Network
- RF Radio Frequency
- Bluetooth Bluetooth
- NFC Near-Field Communication
- satellite broadcasting network analog broadcasting network
- DMB Digital Multimedia Broadcasting
- the term at least one is defined as a term including singular and plural, and even if the term at least one does not exist, each component may exist in singular or plural, and may mean singular or plural. This should be self-explanatory. In addition, whether each component is provided in singular or plural form may be changed depending on the embodiment.
- At least one user terminal 100 sets bedtime and wake-up time using a web page, app page, program, or application related to a real-time sleep health management service using AI-based brain wave entrainment and autonomic nervous system control, and emits light accordingly.
- It may be a user terminal that outputs sound, vibration, and breathing techniques.
- At least one user terminal 100 may be implemented as a computer capable of accessing a remote server or terminal through a network.
- the computer may include, for example, a laptop equipped with a navigation system and a web browser, a desktop, a laptop, etc.
- at least one user terminal 100 may be implemented as a terminal capable of accessing a remote server or terminal through a network.
- At least one user terminal 100 is, for example, a wireless communication device that guarantees portability and mobility, and includes navigation, personal communication system (PCS), global system for mobile communications (GSM), personal digital cellular (PDC), PHS (Personal Handyphone System), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), Wibro (Wireless Broadband Internet) ) It may include all types of handheld-based wireless communication devices such as terminals, smartphones, smartpads, and tablet PCs.
- PCS personal communication system
- GSM global system for mobile communications
- PDC personal digital cellular
- PHS Personal Handyphone System
- PDA Personal Digital Assistant
- IMT International Mobile Telecommunication
- CDMA Code Division Multiple Access
- W-CDMA Wide-Code Division Multiple Access
- Wibro Wireless Broadband Internet
- the management service providing server 300 may be a server that provides a real-time sleep health management service web page, app page, program, or application using AI-based brain wave entrainment and autonomic nervous system control. And, the management service providing server 300 may be a server that databases settings according to light, sound, vibration, and breathing methods. Additionally, the management service providing server 300 may be a server that allows the user terminal 100 to output at least one of light, sound, vibration, and breathing methods for brain wave entrainment and autonomic nervous system control at bedtime and waking time.
- the management service providing server 300 allows bi-lateral somatosensory stimulation (Bi-Lateral Alternating Stimulation in Tactile) to be provided through the user terminal 100 and the wearable device 400, and non-invasive vagus nerve stimulation ( It may be a server that controls the vagus nerve stimulator to relieve pain through NonInvasive Vagus Nerve Stimulation or outputs high frequencies to the user terminal 100.
- bi-lateral somatosensory stimulation Bi-Lateral Alternating Stimulation in Tactile
- non-invasive vagus nerve stimulation It may be a server that controls the vagus nerve stimulator to relieve pain through NonInvasive Vagus Nerve Stimulation or outputs high frequencies to the user terminal 100.
- the management service providing server 300 may be implemented as a computer that can connect to a remote server or terminal through a network.
- the computer may include, for example, a laptop equipped with a navigation system and a web browser, a desktop, a laptop, etc.
- At least one wearable device 400 uploads sensing data that allows the user to determine the sleep stage using a web page, app page, program, or application related to a real-time sleep health management service using AI-based brain wave entrainment and autonomic nervous system control, It may be a device that outputs vibration of a preset frequency according to a preset intensity and time.
- At least one wearable device 400 may be implemented as a computer capable of accessing a remote server or terminal through a network.
- the computer may include, for example, a laptop equipped with a navigation system and a web browser, a desktop, a laptop, etc.
- at least one wearable device 400 may be implemented as a terminal capable of accessing a remote server or terminal through a network.
- At least one wearable device 400 is, for example, a wireless communication device that ensures portability and mobility, and includes navigation, personal communication system (PCS), global system for mobile communications (GSM), personal digital cellular (PDC), PHS (Personal Handyphone System), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), Wibro (Wireless Broadband Internet) ) It may include all types of handheld-based wireless communication devices such as terminals, smartphones, smartpads, and tablet PCs.
- PCS personal communication system
- GSM global system for mobile communications
- PDC personal digital cellular
- PHS Personal Handyphone System
- PDA Personal Digital Assistant
- IMT International Mobile Telecommunication
- CDMA Code Division Multiple Access
- W-CDMA Wide-Code Division Multiple Access
- Wibro Wireless Broadband Internet
- FIG 2 is a block diagram for explaining the management service providing server included in the system of Figure 1, and Figures 3 and 4 are real-time sleep health using AI-based brain wave entrainment and autonomic nervous system control according to an embodiment of the present invention. This is a diagram to explain an embodiment in which a management service is implemented.
- the management service providing server 300 includes a database unit 310, a setting unit 320, a control unit 330, a wake-up brain wave induction unit 340, a sleep stage coaching unit 350, and an artificial intelligence unit. It may include (360), an insomnia excessive relief unit (370), a sunlight shower unit (380), a circadian rhythm adjustment unit (390), a vagus nerve stimulation unit (391), and an alternating somatosensory stimulation unit (393).
- the management service providing server 300 according to an embodiment of the present invention or another server (not shown) operating in conjunction with at least one user terminal 100 and at least one wearable device 400 performs AI-based brain wave entrainment and When transmitting a real-time sleep health management service application, program, app page, web page, etc. using autonomic nervous system regulation, at least one user terminal 100, at least one wearable device 400, and at least one information providing server ( 500) can install or open real-time sleep health management service applications, programs, app pages, web pages, etc. using AI-based brain wave tuning and autonomic nervous system control. Additionally, a service program may be run on at least one user terminal 100, at least one wearable device 400, and at least one information provision server 500 using a script executed in a web browser.
- a web browser is a program that allows the use of web (WWW: World Wide Web) services and refers to a program that receives and displays hypertext written in HTML (Hyper Text Mark-up Language), for example, Chrome. , Microsoft Edge, Safari, FireFox, Whale, UC Browser, etc.
- HTML Hyper Text Mark-up Language
- an application refers to an application on a terminal and includes, for example, an app running on a mobile terminal (smartphone).
- the database unit 310 can store settings for light, sound, vibration, and breathing methods for brain wave entrainment and autonomic nervous system control at bedtime and waking time.
- brain waves are biometric information measured due to constructive interference in the microcurrents of nerve cells when the brain is activated. It measures continuous data at one location and has amplitude, frequency, and waveform, which are characteristics of a wave. Because brain waves show different characteristics depending on the state of nerve cell activation, brain activity can be measured using brain waves. Therefore, polysomnography evaluates the depth of sleep using brain waves. Because people's physical activity decreases unconsciously in a sleep state, sleep EEGs measured in a sleep state show different characteristics from EEGs measured in an awake (non-sleep) state.
- the sleep stage which is the depth of sleep
- the sleep recording manual Rechtschaffen & Kales Sleep Scoring Manual
- This sleep stage standard is used in various research fields, including polysomnography, and is shown in Table 1 below.
- step frequency characteristics Other Features wake up Alpha waves (8-13Hz) dominate Noise caused by movement and heart rate
- the wake stage is a non-sleep state, where alpha waves (8-12 Hz) are activated and noise is generated due to movement and heart rate.
- stage 1 sleep mixed brain waves of 2-7Hz are measured, and K complexes and sleep spindle waves are not detected. K-complexes are transient, high-amplitude, positive-directed waves, and sleep spindles are short, 12-14 Hz complex waves that follow the K complexes.
- stage 2 sleep K complexes and sleep spindle waves are measured, and slow waves of 2 Hz or less make up less than 20%.
- Stage 3 sleep is classified into three stages when 20% to 50% of the brain waves are slower than 2Hz.
- Stage 4 sleep includes more than 50% of slow brain waves below 2Hz.
- REM sleep is similar to stage 1 sleep, but noise caused by eye movements is also measured. At this time, the types and frequencies of brain waves are shown in Table 2 below.
- Types of brain waves frequency band condition of the brain Delta 0.5 ⁇ 4 Hz deep sleep state Theta 4 ⁇ 7 Hz Sleepy state, distracted state, daydreaming state Alpha 8 ⁇ 12Hz A state of relaxed external concentration Sensory Motor Rhythm (SMR) 12 ⁇ 15Hz Maintain concentration without moving A state between tension and relaxation Beta 15 ⁇ 30Hz Stay focused while thinking and being active Gamma 31 ⁇ 50Hz Information exchange between cortical and subcortical areas Appears in dreams during conscious awakening and REM sleep May overlap with beta waves
- Brain wave tuning theory (The Frequence-Following Effect) is also called neural entrainment based on electroencephalogram synchronization, and is caused by brain waves (large electrical oscillations in the brain) caused by the rhythm of periodic external stimuli such as flashing lights, voices, music, and music. ) is adjusted to the desired frequency. This is V. A. Korshunov, G. R. Khazankin and D. S.
- the sympathetic nerve In the case of the sympathetic nerve, it comes out of the middle part of the spinal cord and is distributed to various internal organs, and plays a role in helping people respond quickly in case of an emergency.
- sympathetic nerves When sympathetic nerves are excited, pupils dilate, sweat secretion is promoted, heart rate increases, blood vessels constrict, bronchi dilate, and gastrointestinal motility decreases.
- the parasympathetic nerves come out of the midbrain, medulla, and tail of the spinal cord, are distributed to each internal organ, and play the role of storing energy in preparation for emergency situations.
- the sympathetic nerve induces awakening and, conversely, the parasympathetic nerve induces sleep. Therefore, when inducing sleep, red and yellow light that induces melatonin is provided, and breathing is induced into a deep breathing method as shown in Figures 4e and 4f. Helps stimulate the parasympathetic nervous system.
- red and yellow light induces melatonin is found in the paper (Blume C, Garbazza C, Spitschan M. Effects of light on human circadian rhythms, sleep and mood. Somnologie (Berl). 2019 Sep;23(3) :147-156. doi: 10.1007/s11818-019-00215-x. Epub 2019 Aug 20. PMID: 31534436; PMCID: PMC6751071.) and based on Fig. 4d.
- Upon awakening a situation opposite to the method described above can be induced.
- the display of the user terminal 100 can be gradually brightened before the wake-up time, and the display and flash light can be turned on at maximum brightness as the alarm time approaches.
- alpha-band or beta-band flicker from an LED device, display, or flash light, that is, an image frame that can give a flickering effect to video content.
- Flicker can also be caused by adjusting the display refresh rate and blinking the display, flash light, and associated LEDs. Accordingly, flicker can be used to synchronize alpha waves or beta waves. This theory is based on the above-mentioned paper.
- brain wave changes can be induced by coaching breathing methods using changes in the size of shapes and numbers through the display of the user terminal 100.
- This evidence is based on the paper (Seungwan Kang (2017). On the mechanism by which conscious breathing affects the autonomic nervous system and brain waves. Perspectives in Nursing Science, 14(2), 64-69.).
- the autonomic nervous system such as heart rate
- the autonomic nervous system can be controlled through conscious breathing changes. For example, this is why a person who is very angry and whose sympathetic nervous system is extremely excited is asked to take deep breaths. The idea is to consciously change your heart rate through breathing.
- a user interface for controlling breathing methods for each user can be provided as shown in Figure 4f.
- the ratio of each breathing step can be maintained, but the speed, number of times, and sets can be adjusted according to the level of proficiency in the breathing method to select a breathing method suitable for each user.
- This breathing technique is relatively safe, but you may feel a little dizzy when you first practice it.
- Normal breathing is a balance between inhaling oxygen and exhaling carbon dioxide, but when this balance is disrupted by exhaling more than inhaling, it results in a decrease in carbon dioxide in the body. This causes the blood vessels that supply blood to the brain to narrow, and the decrease in blood supply can lead to symptoms such as dizziness.
- the display can change from a dark black color to a red color that gradually becomes brighter as you breathe in.
- Figure 4g [Inhale (the display gradually changes from black at minimum brightness to red at maximum brightness) ⁇ Hold breath (display blinks) ⁇ Exhale (screen brightness and color gradually darken until it reaches minimum brightness). You can also set it to [Switch to] ⁇ Rest between sets (maintain display brightness and color)].
- the setting unit 320 may receive settings for bedtime and wake-up time from the user terminal 100 .
- the user terminal 100 can set bedtime and wake-up time. As shown in Figure 4a, you can set an alarm for bedtime and wake-up time. A 12-hour or 24-hour clock is provided in graphic form, and you can set the bedtime by moving the moon icon, and use the sun or alarm clock icon. You can set the wake-up time by moving it.
- the difference between bedtime and wake-up time can be calculated to inform the expected sleep time. As the expected sleep time approaches the recommended sleep time, for example, 8 hours, water is filled in the clock shape. It can be.
- the user terminal 100 may be a smartphone, but as described above, it can provide a wake-up alarm using light and brain wave tuning from a display such as a smart pad, smart watch, PC, TV, or headset.
- the control unit 330 may output at least one of light, sound, vibration, and breathing methods for brain wave entrainment and autonomic nervous system control at the bedtime and wake-up time set in the user terminal 100.
- the user terminal 100 may irradiate light of a melatonin-inhibiting wavelength while outputting a sleep induction method, which is a breathing method that stimulates the parasympathetic nervous system at bedtime.
- a sleep induction method which is a breathing method that stimulates the parasympathetic nervous system at bedtime.
- light of the melatonin non-suppressing wavelength may include red and yellow light.
- the user terminal 100 inserts and outputs an alpha-band or beta-band flicker to synchronize alpha or beta brain waves at the time of waking up, and outputs blue light.
- Awakening can be induced by printing.
- blue light can also induce awakening using blue light using a blue light bandpass filter that can be attached to an LED device, display, or flash light that is linked to application software.
- the theory that blue light promotes awakening is based on the paper that it promotes awakening in anesthesia (Liu D, Li J, Wu J, Dai J, Chen X, Huang Y, Zhang S, Tian B, Mei W. Monochromatic Blue Light Activates Suprachiasmatic Nucleus Neuronal Activity and Promotes Arousal in Mice Under Sevoflurane Anesthesia. Front Neural Circuits. 2020 Aug 18;14:55. doi: 10.3389/fncir.2020.00055.
- the wake-up brain wave induction unit 340 generates a monaural beat or binaural beat for brain wave entrainment of theta waves, alpha waves, and beta waves through the speaker of the user terminal 100 from before a preset time from the wake-up time. Beats) sound can be output.
- theta-alpha-beta higher frequency
- brain wave entrainment of theta, alpha, and beta waves can be sequentially induced so that one can gradually wake up and wake up. Since this brain wave entrainment theory is the same as the brain wave entrainment theory described above, detailed explanation will be omitted.
- the sleep stage coaching unit 350 may monitor the sleep stage in the wearable device 400 linked to the user terminal 100 from bedtime and output a preset monaural beat or binaural beat for each sleep stage. At this time, the sleep stage coaching unit 350 may output binaural beats when earphones or headphones are linked to the user terminal 100 and output monaural beats when connected to a speaker.
- the artificial intelligence unit 360 can determine the sleep stage by inputting the collected data collected from the wearable device 400 as a query to a pre-built artificial intelligence algorithm.
- the sleep stage can be determined based on GPS, acceleration sensor, geomagnetic sensor, etc. collected by the wearable device 400, how much the person tosses and turns, what the heart rate is, etc.
- the human sleep cycle starts from WAKE and largely consists of repetitions of the NREM and REM sleep stages, and biosignals measured during sleep have different characteristics for each sleep stage.
- WAKE stage muscles are active and breathing and heartbeat are irregular.
- NREM stage heart rate, breathing, and eye movements slow down compared to the WAKE stage.
- you enter the REM stage after the NREM sleep stage your heart rate and breathing become fast and irregular again, and the tension in the muscles below the neck decreases.
- Respiration rate is affected by the user's movements and noise is generated. Therefore, data that is judged to be missing because the respiratory rate is 0 or the value is too large or too small at a specific point can be corrected with the average data of the preceding and following values.
- breathing is characterized by irregularity. Reflecting the characteristic that the amplitude increases as breathing becomes irregular, the maximum amplitude of a window with a size of 3 minutes can be measured and that value can be reflected as a feature.
- the heart rate stabilizes the moment one enters the NREM sleep stage, and in the REM sleep stage, the heart rate graph shows an increasing and decreasing trend, and the amplitude appears larger than in other sleep stages. Reflecting these characteristics, the magnitude of the amplitude can be extracted through a window of 1 minute in size.
- QRS Complex can be extracted according to the Pan-Tomkins algorithm, which is commonly used in electrocardiogram (ECG) analysis.
- ECG electrocardiogram
- HRV Heart Rate Variability
- amplitude size amplitude size
- number of R wave peaks per minute Heart rate variability analysis is based on R-Peak detection, which is the biggest feature of the QRS Complex.
- Heart rate variability shows different characteristics depending on the sleep stage, and can distinguish between awakening and sleeping states with an accuracy of over 87%.
- Equation 1 and Equation 2 the standard deviation of the R-R PEAK interval is obtained at 1-minute intervals, and the Short Term SDNN (Standard Deviation of NNInterval), which is a short-term heart rate variability factor, is obtained.
- Data collected from the 3-axis acceleration sensor is used to determine movement during sleep. Since there is no need to consider directionality, the three-dimensional acceleration sensor value can be reduced to a one-dimensional value and converted into an intensity value expressing the intensity of movement during sleep. By substituting the acceleration at point n on the XYZ axis into Equation 3 below, In, the movement intensity at point n, is obtained.
- Clock Proxy that represents the sleep cycle instead of the absolute time corresponding to the measured biological signal
- the existing biological clock modeling technique can be used to assign a value from the start to the end of the measurement time.
- Raw data collected by each sensor can be visualized to read the waveforms that appear in each sleep stage, and then compared to the waveforms observed in previous studies, labels of 0 for WAKE state, 1 for NREM, and 2 for REM can be assigned. there is.
- the measurement time can be integrated into one data frame and saved as a CSV file.
- the CSV file can include heart rate variability, breathing variability, movement intensity, and clock proxy as features. Data that has completed the preprocessing process can be used as training and testing data for the SVM classifier.
- the SVM classifier obtains a hyperplane that maximizes the distance between the decision hyperplane that separates each class and the closest sample (Support Vector), and has excellent generalization ability in classification problems. Therefore, the impact of error data is small and there are few cases of overfitting.
- the Python Scikit-Learn library sleep data that cannot be linearly classified can be classified by matching them to a higher dimension than the finite dimension of the initial problem using the RBF (Radial Basis Function) kernel.
- the accuracy of learning can be increased by selecting the OvO (One-verses-One) method, although it takes longer processing time than the OvR (One-verses-Rest) method.
- the two RBF kernel parameters, C and ⁇ can be specified as 1 and 0.1, respectively, through an empirical method using Grid Search.
- sleep efficiency which is the ratio of each sleep stage to total sleep and the ratio of actual sleep time among the recorded time
- the Python Matplotlib library can be used to visualize the sleep stage for the entire sleep time as a sleep curve (Hypnogram) and the ratio of each sleep stage with a pie chart.
- Sleep curves can be used to identify sleep stages, and it is possible to prepare for bed and wake up according to the sleep stage.
- the excessive insomnia alleviation unit 370 uses monaural beats or binaural beats to synchronize alpha waves, theta waves, and delta waves by manual setting or by sleep stage. It provides beats and can stop brain wave entrainment after the deep sleep stage to induce REM sleep. If hypersomnia is registered in the user terminal 100 or not registered as hypersomnia, the insomnia hypersomnia alleviation unit 370 uses monaural beats or binaural beats to synchronize by manual setting or in the order of theta waves, alpha waves, and beta waves. Lalbit can be provided.
- the case of not registering as insomnia means that brain wave entrainment is induced manually or automatically (sleep stage), such as in cases where insomnia is not registered but the patient cannot fall asleep, as there are cases where the patient cannot fall asleep even if the patient is not registered as insomnia.
- Cases that are not registered as hypersomnia mean that sometimes there are cases where you sleep too much even if you are not registered as hypersomnia. Therefore, in cases where you sleep too much but are not hypersomnia, it is recommended to induce brain wave synchronization manually or automatically (sleep stage). it means.
- the excessive insomnia alleviation unit 370 may output binaural beats when earphones or headphones are linked to the user terminal 100 and output monaural beats when connected to a speaker.
- insomnia or hypersomnia can help treat circadian rhythm disorder, bipolar disorder, depression, manic depression, anxiety disorder, Alzheimer's disease, and Parkinson's disease, and in the case of jet lag for travelers and shift workers, the circadian rhythm disorder. It can help you re-establish your rhythm.
- the golden 90-minute sleep rule was introduced in the best-selling book “Stanford High-Efficiency Sleep Method” by Dr. Seiji Nishino, director of the Sleep and Circadian Neurobiology Laboratory at Stanford University. According to the "90-minute rule of time,” as shown in Figure 4c, the sleep quality of the first sleep cycle immediately after waking up has a significant impact on the overall sleep quality.
- non-REM sleep occurs for 70 to 90 minutes.
- Non-REM sleep is a deep sleep stage that relieves fatigue and stores memories.
- the 90 minutes during which the first non-REM sleep occurs is called the [golden time of sleep]. If you sleep well for 90 minutes right after falling asleep, you can feel refreshed the next day even if you sleep less than usual.
- sleep pressure is released the most during the 90 minutes after falling asleep, that is, during the golden time of sleep. If you get a good night's sleep during the golden sleep time, sleep pressure is greatly reduced, the desire to sleep is relieved, and fatigue is reduced. Sleep golden time regulates the autonomic nervous system through sleep, promotes the secretion of growth hormones, and improves brain condition.
- a key role of REM sleep is to regulate emotional reactivity and emotional information processing in depression.
- Neurometabolic changes in depression are triggered or enhanced by REM sleep hyperactivation.
- the time of the deep sleep stage of the first sleep cycle can be increased using delta wave brain wave entrainment. By monitoring sleep stages, you can disable REM sleep hyperactivity by waking up in the REM sleep stage.
- REM rebound (shortened REM latency and increased REM density) caused by REM instability can cause depression.
- Brain waves can also be synchronized through monaural beats or binaural beats using inaudible frequencies.
- the sunlight shower unit 380 can provide blue light, which is blue light, according to the waking time.
- the sunlight shower alarm and light therapy time can be provided as shown in Figure 4h.
- Effective usage timing and usage amount are predicted by an algorithm through user data, and the clock shape can be filled as the recommended usage timing and usage amount approaches. It can tell you the optimal time to take a morning shower by taking into account the user's sleep pattern. In this case, the sleep pattern is when you go to sleep and when you wake up. Or, it can tell you when to take a morning shower depending on when you should sleep and when you should wake up. The best thing for insomnia or oversleeping is to fix the waking time and see the sunlight as soon as you wake up.
- the sleep hormone melatonin When exposed to morning sunlight rich in blue light, the sleep hormone melatonin is suppressed in the morning, the serotonin hormone is secreted, and the serotonin hormone is converted to melatonin after 14 to 15 hours to induce sound sleep. You can guide the time to take a sunlight shower or do light therapy 14 to 15 hours before the set or predicted bedtime and notify with an alarm.
- the circadian rhythm matching unit 390 may emit blue light based on the circadian rhythm determined by user data of the user terminal 100 in addition to the wake-up time.
- sleep data, light exposure data, and activity data such as sleep stage, sleep time, bedtime, wake-up time, light exposure time of the illuminance sensor, light exposure time, GPS data, and heart rate data can be measured.
- circadian rhythm can be measured using a preset algorithm, circadian rhythm disorders can be diagnosed, and personalized light therapy can be provided through circadian rhythm data analysis. In other words, the exposure point, exposure intensity, and exposure time of blue light can be adjusted. Since the human circadian clock runs on a cycle slightly longer than 24 hours, brain wave entrainment can be used to advance the clock each day.
- the vagus nerve stimulation unit 391 When a circadian rhythm disorder is monitored, the vagus nerve stimulation unit 391 outputs vibration corresponding to a vagus nerve stimulator (NonInvasive Vagus Nerve Stimulation) from the user terminal 100 or connects with the vagus nerve stimulator to generate a vagus nerve stimulator. can be driven.
- the non-invasive vagus nerve stimulator is a portable, user-friendly device (gammaCoreTM) that delivers high-frequency electrical pulses (25 Hz, 2 minutes) to stimulate the afferent fibers of the vagus nerve in the skin of the neck, and specifically to the trigeminal neurovascular system. Regulates the central inflow of nerves.
- ACT-2 study also showed that nVNS treatment was more effective than the placebo treatment group in the episodic cluster headache group in terms of the proportion of headache attacks in which pain disappeared within 15 minutes (48% vs.6%, p ⁇ 0.01). Accordingly, it has proven to have moderate efficacy and excellent tolerability in cluster headache patients, so it can be considered as an acute treatment for cluster headaches that cannot receive standard therapy or do not receive sufficient relief from drugs.
- the theoretical basis for this is in the paper (Silberstein SD, Mechtler LL, Kudrow DB, Calhoun AH, McClure C, Saper JR, Liebler EJ, Rubenstein Engel E, Tepper SJ; ACT1 Study Group.
- Non-Invasive Vagus Nerve Stimulation for the ACute Treatment of Cluster Headache Findings From the Randomized, Double-Blind, Sham-Controlled ACT1 Study. Headache. 2016 Sep;56(8):1317-32. doi: 10.1111/head.12896. PMID: 27593728; PMCID: PMC5113831.) It is based on
- Sensate®Somacoustics which has a thesis (Sensate® Somacoustics: A New Wave for Stress Management. Volume I By, Scott McDoniel, Ph.D., M.Ed.1 & Stefan Chmelik, M.Sc. 2 1 Faculty; College of Health Professions, Walden University, Minneapolis, MN 2. BioSelf Technologies, Ltd, London, UK).
- the U.S. Food and Drug Administration has approved Vagus Nerve Stimulation as a long-term adjunctive treatment for chronic recurrent depression in patients 18 years of age and older. At this time, whole body vibration between 20 and 30 Hz significantly increases the secretion level of the brain and increases proteins for necessary neural plasticity.
- Sensate® is a non-electrical oscillatory device for stress management that uses low-frequency technology ( ⁇ 50 Hz) to improve parasympathetic nervous system (PNS) response to chronic stress.
- the two theoretical concepts of the Sensate® device are Bone Conduction and Thoracic Resonance.
- the sternum can lead to greater amplification due to bone movement and additional energy can come from resonance properties within the chest.
- Low-frequency sound waves (60 Hz) travel longer distances around internal organs, and the afferent nerve branches leading to the vagus nerve are located throughout the chest cavity. Therefore, the Sensate® device is a new vagus nerve stimulation method that enhances parasympathetic nervous system function.
- An increased parasympathetic nervous system can improve anxiety, insomnia, and other stress-related symptoms.
- the alternating somatosensory stimulation unit 393 moves the user terminal 100 to one hand or wrist of the user.
- vibration with a preset frequency can be output.
- the frequency, intensity, duration, and number of vibrations can be set to increase or decrease in the user terminal 100.
- Bi-Lateral Alternating Stimulation in Tactile may provide a non-invasive, non-pharmacological means of managing stress.
- alternating bilateral physical sensory stimulation can be effective in reducing subjective stress and anxiety levels and has a beneficial effect on patients with anxiety.
- BLAST has a depotentiating effect on synapses in the amygdala that are activated during recall of fear-based memories. These results suggest that BLAST may affect electrical activity in key brain regions associated with stress and anxiety, with the overall effect being to alleviate the stress response in humans and reduce or eliminate painful flashbacks or physical sensations associated with physical pain. It suggests that there is. BLAST can reduce sympathetic activation by reducing electrical activity in key regions of the salience network. Additionally, since people with OCD may have severe anxiety that limits their attention and creates hypervigilance to internal and external distractions, they may see a significant reduction in hyperactivity using BLAST.
- circadian rhythm dysregulation is bipolar disorder. It is closely related to circadian rhythm, and management of circadian rhythm is of great help in improving the prognosis of bipolar disorder, and it is necessary to develop effective treatment methods for this.
- brain wave entrainment can be performed by activating beta waves and balancing the brain wave frequencies of the left and right brains.
- the goal is to balance alpha waves by decreasing them in areas of the left hemisphere of the brain and increasing them in the right hemisphere.
- monaural beats to synchronize the alpha waves of the right brain are played through the left ear.
- beta wave synchronized monaural beats or binaural beats When performing blue light therapy, play beta wave synchronized monaural beats or binaural beats.
- Coaching deep breathing techniques stimulates the parasympathetic nervous system and strengthens alpha brain waves.
- gamma brain waves can be induced by playing monaural beats or binaural beats of the gamma band.
- FIGS. 4i to 4n analyzing the brain waves (quantitative brain waves, QEEG) of Figures 4i to 4n is shown in Table 4 below.
- the source of FIGS. 4i to 4n is iMediSync, as described above.
- Each disease or symptom can be alleviated by entraining to compensate for excessive or insufficient brain waves.
- FIG 4i mild cognitive impairment -In the EEG of normal aging healthy people, normal EEG waves such as alpha waves are observed (yellow) - In the EEG of patients with mild cognitive impairment, slow EEG waves such as theta waves are observed (green) - In the EEG of Alzheimer's dementia patients, delta waves and theta waves are observed. Very slow brain waves such as (blue) are observed (blue).
- FIGS. 3 and 4 as an example.
- the embodiment is only one of various embodiments of the present invention and is not limited thereto.
- the stress index can be lowered through alternating sensory stimulation, and chronic pain can be alleviated through vagus nerve stimulation, as shown in (d).
- the diseases whose symptoms can be alleviated through various brain wave entrainment are listed in Table 3, so duplicate explanations will be omitted.
- FIG. 4 since FIG. 4 has been fully described in FIG. 2, duplicate description thereof will also be omitted.
- FIG. 5 is a diagram illustrating a process in which data is transmitted and received between components included in the real-time sleep health management service providing system using AI-based brain wave tuning and autonomic nervous system control of FIG. 1 according to an embodiment of the present invention.
- FIG. 5 an example of the process of transmitting and receiving data between each component will be described with reference to FIG. 5, but the present application is not limited to this embodiment, and the process shown in FIG. 5 according to the various embodiments described above It is obvious to those skilled in the art that the process of transmitting and receiving data can be changed.
- the management service providing server may store settings for light, sound, vibration, and breathing methods for brain wave entrainment and autonomic nervous system control at bedtime and wake-up time.
- the management service providing server receives the bedtime and wake-up time set from the user terminal (S5200), and selects light, sound, vibration and breathing methods for brain wave entrainment and autonomic nervous system control at the bedtime and wake-up time set by the user terminal. At least one can be output (S5300).
- Figure 7 is a diagram showing the configuration of a real-time sleep health management service providing system 11 using AI-based brain wave tuning and autonomic nervous system control according to an embodiment of the present invention.
- the real-time sleep health management service providing system 11 using AI-based brain wave tuning and autonomic nervous system control will be referred to as the present system 11 for convenience of explanation.
- this system 11 i.e., a real-time sleep health management service providing system using AI-based brain wave entrainment and autonomic nervous system control
- a smart watch 1100 may be configured to include a smart watch 1100 and an analysis server 1200. there is.
- This system 11 refers to a system that coaches the user's sleep using the user's biometric data collected (obtained) using the smart watch 1100.
- This system 11 can provide sleep coaching technology to improve the user's sleep quality through analysis of the user's biometric data available through the smart watch 1100.
- the smart watch 1100 is in close contact with the user's body and can collect sleep biometric data including the user's heart rate, movement, and blood pressure during sleep.
- devices that collect a user's sleep biometric data may include various wearable devices such as a smart watch. That is, in the present invention, the smart watch 1100 may refer to a wearable device that is worn on a part of the user's body and collects the user's sleep biometric data. In the present invention, the smart watch 1100 may be applied not only to a smart watch but also to various types (forms) of wearable devices (wireless communication devices).
- the smart watch 1100 may collect sleep biometric data of a user in a sleeping state (that is, in a state in which the user is sleeping and is not moving).
- the smart watch 1100 may refer to a wearable device owned by a user.
- the movement in the sleep biometric data may mean, for example, the movement of the user's body corresponding to the movement (motion) data of the smart watch 1100, which is a motion detection sensor built into the smart watch 1100.
- the motion detection sensor may include a 3-axis accelerometer, 3-axis gyroscope, 3-axis magnetometer, etc.
- the analysis server 1200 may link a sleep monitoring algorithm with the smart watch 1100 to provide the user with an alarm for personalized sleep coaching tailored to each user.
- the analysis server 1200 may be referred to differently as, for example, a real-time sleep coaching server (device) using artificial intelligence.
- the analysis server 1200 provides web pages, app pages, programs, or applications related to real-time sleep coaching using artificial intelligence to a user terminal (not shown) or smart watch 1100 owned by a user using the analysis server 1200. This allows users to receive notifications for personalized sleep coaching.
- the user terminal (not shown) is a portable terminal carried by the user and may mean, for example, a smartphone, smart pad, tablet PC, laptop, etc.
- the smart watch 1100 is a type of user terminal and may refer to a wearable device.
- the analysis server 1200 may collect biometric sleep data for each user from the smart watch 1100 and accumulate biometric sleep data depending on whether the user moves. For example, the analysis server 1200 may accumulate sleep biometric data, which is the heart rate while the user is not moving. Additionally, the analysis server 1200 can measure biometric data while the user moves.
- the analysis server 1200 learns and analyzes the accumulated sleep biometric data with an artificial intelligence model (i.e., deep learning model) to identify sleep patterns and optimal sleep cycle information for each user, and determines wake-up time based on the identified optimal sleep cycle information. can be calculated.
- the analysis server 1200 compares the previously stored sleep cycle data and optimal sleep cycle information with the sleep cycle calculated through sleep biometric data for each user, predicts the user's sleep cycle according to the comparison result, and provides optimal sleep cycle information and optimal sleep. Wake-up time can be calculated according to the cycle.
- the smart watch 1100 receives wake-up time information (i.e., wake-up time information, which is information about the wake-up time calculated by the analysis server) from the analysis server 1200, and sends a wake up alarm (i.e., wake-up time) at the wake-up time.
- wake-up time information i.e., wake-up time information, which is information about the wake-up time calculated by the analysis server
- a wake up alarm i.e., wake-up time
- Alarm, wake-up notification are provided.
- the analysis server 1200 may accumulate sleep biometric data for each user collected while the user is sleeping and not moving through a wearable device such as a smart watch 1100. Afterwards, the analysis server 1200 can analyze sleep patterns for each user using the accumulated sleep biometric data using an artificial intelligence model and identify optimal sleep cycle information for each user.
- the optimal sleep cycle information for each user is a generalization of information about what REM sleep period the user feels most refreshed upon waking up, and the analysis server 1200 uses the optimal sleep cycle information for each user based on the biometric information collected according to the waking time. Through analysis of information, you can learn about each user.
- the user's biometric data (which may be otherwise referred to as biometric information) collected by the analysis server 1200 from the smart watch 1100 is, for example, i) when the user is sleeping (or not waking up) It is divided into ii) biometric data (biometric information) measured when the user is awake from sleep (or active, awake). It can be.
- biometric data corresponding to case i) may be differently referred to in the present invention as sleep biometric data
- the biometric data corresponding to case ii) may be referred to by the term user biometric information (or non-sleep biometric data) in the present invention. It may be referred to differently.
- biometric data may mean data (information) such as heart rate (heart rate), blood pressure, and movement.
- the user's state can be largely divided into a 'sleeping state (i.e., a sleeping state)' and a 'non-sleeping state', where the 'non-sleeping state' refers to a state that is not sleeping in the present invention. It may be referred to differently as a state (non-sleep state), an active state, a waking state, a waking state (awake state), a normal state, etc.
- the analysis server 1200 uses wearable devices such as a smart watch 1100 to classify the user's presence or absence of movement, accumulates heart rate data when there is no movement, and predicts and classifies the user's sleep stage through deep learning of the accumulated data. can do. Afterwards, the analysis server 1200 determines the REM sleep stage by considering the sleep stage and heart rate change, and provides a wake-up alarm through the smart watch 1100 in the light sleep stage after REM sleep. . Additionally, the analysis server 1200 may repeat the alarm until it confirms that the user is awake through the user's movements and heart rate. In other words, the analysis server 1200 can repeatedly generate a weather alarm and repeatedly provide it to the user through the smart watch 1100 until the user confirms that he or she is in a good weather state.
- wearable devices such as a smart watch 1100 to classify the user's presence or absence of movement, accumulates heart rate data when there is no movement, and predicts and classifies the user's sleep stage through deep learning of the accumulated data. can do
- the analysis server 1200 analyzes biometric data collected through the smart watch 1100 (for example, data such as user movement and heart rate corresponding to the motion of the smart watch) using a deep learning model to determine the user's current sleep stage. can be classified and predicted.
- biometric data collected through the smart watch 1100 for example, data such as user movement and heart rate corresponding to the motion of the smart watch
- a deep learning model to determine the user's current sleep stage. can be classified and predicted.
- the user's sleep stage considered by the analysis server 1200 may include a REM sleep stage and a non-RAM sleep stage.
- the non-RAM sleep stage may include a light sleep stage and a deep sleep stage.
- the REM sleep stage is a sleep stage located between the waking state (awakening state) and the light sleep stage. In the REM sleep stage, the user may dream.
- the light sleep stage is a sleep stage located between the REM sleep stage and the deep sleep stage, and may include, for example, a stage 1 sleep stage in which light sleep occurs and a stage 2 sleep stage in which light sleep occurs.
- the deep sleep stage refers to a sleep stage in which you sleep more deeply than the light sleep stage.
- stage 3 sleep may include stage 3 sleep, a sleep stage in which slow waves appear, and stage 4 sleep, a deep sleep stage in which slow waves occur.
- Stage 3 and stage 4 sleep can be referred to as slow-wave sleep stages.
- Stage 4 sleep can refer to a state of complete deep sleep in which the person does not wake up even if woken up or is not aware of being picked up.
- the analysis server 1200 determines that if the time taken to sleep is different from the usual pattern, the total sleep time will change, so if the user starts sleeping at a time different from the usual pattern, it takes into account the time remaining until the time the user has to wake up and sleeps for 1.5 hours. It adjusts the optimal wake-up time (i.e. optimal wake-up time) suitable for the user's repeated sleep cycle to wake the user.
- the analysis server 1200 provides bedtime (optimal bedtime) and wake-up time (optimal wake-up time) that optimizes sleep efficiency, which is the actual sleeping time compared to the time lying in bed, through sleep biometric data for each user learned through machine learning. You can calculate and set this as the bedtime alarm time and wake-up alarm time.
- the analysis server 1200 performs machine learning based on a deep learning neural network based on a customized training data set optimized for determining sleep stage and identifying optimal cycle information (optimal sleep cycle information) to determine optimal wake-up time.
- a time calculation model can be implemented, and the optimal wake-up time can be calculated using the implemented optimal wake-up time calculation model.
- the analysis server 1200 can be linked with the smart watch 1100 through a network to transmit and receive data.
- the network includes, for example, 3rd Generation Partnership Project (3GPP) network, Long Term Evolution (LTE) network, World Interoperability for Microwave Access (WIMAX) network, Internet, Local Area Network (LAN), and Wireless Local (Wireless LAN).
- 3GPP 3rd Generation Partnership Project
- LTE Long Term Evolution
- WIMAX World Interoperability for Microwave Access
- LAN Local Area Network
- Wireless Local Wireless Local
- Area Network Wide Area Network
- WAN Wide Area Network
- PAN Personal Area Network
- Bluetooth Wireless Local
- NFC Near Field Communication
- satellite broadcasting network satellite broadcasting network
- analog broadcasting network DMB (Digital Multimedia Broadcasting) network, etc.
- DMB Digital Multimedia Broadcasting
- it is not limited to this and may include various wired/wireless communication networks.
- Figure 8 is a diagram showing the data processing configuration of the smart watch 1100 according to an embodiment of the present invention.
- the smart watch 1100 may be configured to include a biometric data collection module 1110, a communication module 1120, a sleep information provision module 1130, and an alarm module 1140.
- module' used in this specification should be interpreted to include software, hardware, or a combination thereof, depending on the context in which the term is used.
- software may be machine language, firmware, embedded code, and application software.
- hardware may be a circuit, processor, computer, integrated circuit, integrated circuit core, sensor, MicroElectro-Mechanical System (MEMS), passive device, or a combination thereof.
- MEMS MicroElectro-Mechanical System
- the term 'module' may also be referred to as 'unit', etc.
- the biometric data collection module 1110 collects sleep biometric data including the user's heart rate, movement, and blood pressure while sleeping, and biometric data including heart rate, movement, and blood pressure even when the user is in an active state waking up from sleep (i.e., the user biometric information) can be collected.
- the communication module 1120 transmits the collected sleep biometric data and biometric data (user biometric information) to the analysis server 1200, and receives wake-up time information from the analysis server 1200.
- the communication module 1120 communicates data with the analysis server 1200 and an external server to provide the user with the user's optimal waking time and sleep analysis information.
- the sleep information providing module 1130 provides the user with sleep pattern information including the user's average sleep time, sleep start time, and wake-up time, and provides the user with daily sleep time information and insufficient sleep time information.
- the sleep information providing module 1130 provides the user's sleep pattern and sleep pattern change information using visual objects such as graphs, icons, and images, so that the user can understand his/her sleep information more intuitively through visual objects. Let it happen.
- the sleep information providing module 1130 can provide such various information to the user by displaying it on the screen (display unit) of the smart watch 1100.
- the sleep information providing module 1130 guides the user on the time at which the user can take a nap or a short nap, and determines when the user's fatigue is at a certain level. In this case, this can be notified to the user.
- the alarm module 1140 may provide a user wake-up alarm at the wake-up time received from the analysis server 1200. That is, the alarm module 1140 may provide a wake-up alarm at a time corresponding to the wake-up time information received from the analysis server 1200.
- the wake-up alarm may be provided to the user in the form of vibration, sound, light, etc., for example.
- Figure 9 is a diagram showing the data processing configuration of the analysis server 1200 according to an embodiment of the present invention.
- the analysis server 1200 includes a data collection module 1210, a prediction module 1220, a calculation module 1230, a weather condition confirmation module 1240, and a re-alarm generation module 1250. It can be.
- the data collection module 1210 collects sleep biometric data including the user's heart rate, movement, and blood pressure during sleep from the smart watch 1100, classifies whether the user moves according to the sleep biometric data, and detects when there is no movement. Accumulate user heart rate data.
- the prediction module 1220 analyzes the accumulated data through deep learning to predict the user's sleep cycle and real-time sleep stage according to the user's biological signals.
- the prediction module 1220 uses a deep learning neural network including at least one of a Deep Neural Network (DNN), a Convolutional Neural Network (CNN), a Recurrent Neural Network (RNN), and a Bidirectional Recurrent Deep Neural Network (BRDNN) as training data.
- DNN Deep Neural Network
- CNN Convolutional Neural Network
- RNN Recurrent Neural Network
- BBDNN Bidirectional Recurrent Deep Neural Network
- Deep learning models considered in the present invention include deep learning, artificial intelligence (AI) algorithm model, artificial neural network, machine learning (machine learning) model, neural network model (artificial neural network model), neuro fuzzy model, and deep learning neural network. It may be referred to differently by terms such as:
- deep learning models include, for example, convolution neural networks (CNN), recurrent neural networks (RNN), and deep neural networks, which are already known or developed in the future. Various neural network models can be applied.
- the deep learning model used by the prediction module 1220 when predicting the user's sleep stage is a pre-learned deep learning model, and may be referred to differently by terms such as a pre-learned sleep stage prediction model.
- the calculation module 1230 calculates optimal sleep cycle information for each user based on the most refreshing RAM sleep cycle information upon waking up pre-stored in the analysis server 1200. For example, the calculation module 1230 may calculate the wake-up time to wake the user from the light sleep stage after the REM sleep stage according to the calculated optimal sleep cycle information.
- the calculation module 1230 can determine the sleep start time at which the user falls asleep through analysis of biometric information (biometric data) received from the smart watch 1100 and compare the sleep start time with the usual pattern. Thereafter, the calculation module 1230 may readjust the wake-up time by considering the repeated sleep cycle according to the comparison result.
- biometric information biometric data
- the calculation module 1230 considers the remaining time to sleep until you need to wake up and allows the user to sleep in the light sleep stage of the last sleep cycle predicted.
- the wake-up time can be calculated so that you can wake up, and then the calculated wake-up time is transmitted to the smart watch 1100 so that you can receive a wake-up alarm at the corresponding time (calculated wake-up time).
- the analysis server 1200 not only collects and analyzes biometric information (biometric data) such as the user's movements and heart rate using the smart watch 1100, but also acquires sleep habit data for each user through deep learning.
- biometric information biometric data
- the sleep habit data may include the number of times the user tosses and turns, the number of times the user wakes up during sleep, and heart rate change information, and this may be data obtained through analysis and learning using a deep learning model.
- the calculation module 1230 in the analysis server 1200 calculates the difference by comparing the sleep start time of the actual user who goes to sleep on average from the acquired sleep habit data, and calculates the time equal to the calculated difference to calculate the last sleep time.
- the smart watch 1100 can be used to wake the user during the light sleep phase of the cycle. As a specific example, if the user falls asleep 1 hour later or earlier than usual, the calculation module 1230 does not wake up at a predetermined time, but sleeps for 1.5 hours according to the sleep cycle graph as shown in FIG. 6, for example.
- the wake-up time can be calculated so that you can sleep and wake up only after the cycle is repeated a certain number of times (for example, 3 or 4 times).
- the calculation module 1230 sets the wake-up time (wake-up time at which the wake-up alarm is made) so that the user wakes up after 4.5 hours when the sleep cycle is repeated 3 times (3 times) or 6 hours after the sleep cycle is repeated 4 times (4 times). can be calculated.
- the analysis server 1200 analyzes the user's biometric data collected from the smart watch 1100 using a deep learning model to determine the user's sleep habits, such as the number of times the user tosses and turns, the number of times the user wakes up, and changes in heart rate. Data can be derived, and the user's sleep habit data can also be learned with a deep learning model.
- the analysis server 1200 can analyze the user's biometric data collected in real time to derive the user's movement changes and heart rate changes every day, and based on this, determine whether the user is currently in the REM sleep stage and determine whether the user is in the REM sleep stage.
- a wake-up alarm can be provided through the smart watch 1100 according to the calculated wake-up time so that the user can wake up in the light sleep stage after the REM sleep stage. That is, the analysis server 1200 analyzes the user's movements and heart rate changes in real time every day, and determines whether the user's sleep stage is a light sleep stage that occurs around the set wake-up time (i.e., the user's sleep stage is after the REM sleep stage). A function that wakes the user up by providing a wake-up alarm to the smart watch 1100 (at a time when it is determined that the user is in a light sleep stage) can be provided.
- the analysis server 1200 determines which user's total sleep time should wake up, which varies every day. Considering the remaining time until the time point (this means the wake-up time calculated by the analysis server) (i.e., taking into account the remaining time remaining from the current point when the user is sleeping to the time corresponding to the calculated wake-up time) By providing a wake-up alarm in the light sleep stage of the predicted last sleep cycle, the user can wake up from sleep by the wake-up alarm provided through the smart watch 1100 in the light sleep stage.
- the calculation module 1230 calculates bedtime and wake-up time at which sleep efficiency, which is the actual sleeping time compared to the time lying in bed, is above a certain level (above a certain level) through sleep biometric data learned with a deep learning model (machine learning). By calculating the time, you can notify the user of the calculated bedtime and wake-up time.
- sleep efficiency considered in the present invention means the actual sleeping time compared to the time lying in bed
- the analysis server 1200 uses the user's sleep biometric data learned with a deep learning model (machine learning) to Bedtime and wake-up time that optimize sleep efficiency can be calculated (predicted) and the calculated information (i.e., optimal bedtime and optimal wake-up time) can be notified to the smart watch (1100).
- the analysis server 1200 provides the calculated information to the user, allowing the user to go to bed at a bedtime when sleep efficiency is optimal and to wake up at a wake-up time when sleep efficiency is optimal, thereby improving sleep quality. This can be improved.
- the calculation module 1230 calculates and notifies the insufficient sleep time compared to the usual pattern, and uses the insufficient sleep time to nap or take a nap. You can be guided to supplement with .
- the calculation module 1230 determines that the required sleep time per day is 8 hours (or, upon analyzing the user's sleep pattern, the user's usual sleep time is 8 hours on average). For example, the user's sleep time collected in real time is 8 hours. If, as a result of analyzing biometric data, it is determined that the user has only slept 5 hours, the system will guide the user to make up for the 3 hours of sleep time (i.e. 3 hours of make-up sleep time) by taking a nap or napping during the day. You can.
- the calculation module 1230 analyzes the user's sleep biometric data collected in real time from the smart watch 1100, derives today's sleep time, which is the time when the user slept today, and then calculates the previously derived sleep time. If today's sleep time is judged to be less than the required sleep time by comparing the user's required sleep time with today's sleep time, the time corresponding to the difference between the required sleep time and today's sleep time is supplemented with sleep time (or insufficient sleep) time), and then supplemental sleep time guidance information that guides the user to supplement the derived supplemental sleep time with a nap or a short nap can be generated and provided to the smart watch 1100.
- This analysis server 1200 provides supplementary sleep time guidance information to the user through the smart watch 1100, allowing the user to calculate the total amount of daily sleep time (i.e., the total sleep time of the day in which actual sleep was achieved) every day. It can be kept the same.
- the supplementary sleep time guidance information may, for example, be something like, 'You are lacking 3 hours of sleep compared to usual (or required sleep time), so please take a 3-hour nap during the day.
- the required sleep time is, for example, generally set to the recommended sleep time essential for the user's corresponding age, or the user's average sleep derived based on analysis of the user's biometric data collected in real time. Can be set to time.
- the analysis server 1200 uses the smart watch 1100 to accumulate and analyze the user's sleep biometric data even when the user takes a nap (even when the user is taking a nap), and analyzes the user by taking into account the sleep cycle. You can wake it up.
- the analysis server 1200 may be configured to provide a repetitive wake-up alarm by determining whether the user has completely woken up even while the user is taking a nap through analysis of biometric data. That is, the analysis server 1200 can provide a wake-up alarm to the user through the smart watch 1100, and repeatedly provide the wake-up alarm until the user's movements and heart rate are confirmed to be in a wake-up state.
- the analysis server 1200 determines that when the user usually takes a nap, the sleep time changes if the user falls asleep earlier or later than the usual sleep time, so the remaining time until the time to wake up is calculated. Considering the time, the wake-up time can be calculated to occur in the light sleep stage of the predicted last sleep cycle.
- the weather condition check module 1240 receives biometric information (biometric data) including the user's movements and heart rate from the smart watch 1100 in real time, analyzes the biometric information, and determines whether the user's biometric information is in a weather state (weather state). ) can be checked.
- the waking state confirmation module 1240 can check whether the biometric information received in real time is the user's biometric information obtained when the user is in an active state (or waking up state, normal state) after waking up from sleep.
- the weather condition confirmation module 1240 compares the movement speed and heart rate (heart rate) in the normal state of awakening from previously stored sleep with biometric data collected from the user's smart watch 1100 to determine whether the user is in a normal state (weather condition).
- the waking state confirmation module 1240 determines that the user is in a waking state (i.e., waking up state). ) can be judged.
- the re-alarm generation module 1250 is configured to operate at a certain time interval (e.g. A wake-up alarm (weather alarm) can be repeatedly generated at 5-minute intervals, and the generated wake-up alarm can be provided through the smart watch 1100.
- a wake-up alarm weather alarm
- the analysis server 1200 uses artificial intelligence (deep learning model) to analyze biometric data (biometric information) including the user's sleep biometric data and user biometric information (i.e., non-sleep biometric data).
- biometric data biometric information
- user biometric information i.e., non-sleep biometric data
- optimal cycle information optimal sleep cycle information
- the analysis server 1200 performs an out of distribution detection process other than learning to process unlearned patterns in addition to noise response.
- Non-learning distribution data detection is to identify whether the data input to artificial intelligence is learned probability distribution data.
- the analysis server 1200 can improve stability and reliability by filtering out images that are difficult for the artificial neural network to judge or processing exceptions through detection of distribution data other than learning.
- the analysis server 1200 may calibrate a probability value indicating how confident it is in the deep learning decision in order to detect distribution data outside of learning, or use a generative adversarial neural network (GAN) to detect distribution data outside of learning.
- GAN generative adversarial neural network
- Adversarial Network Adversarial Network
- the analysis server 1200 allows the user's wake-up time to be finalized using lightweight deep learning technology that simplifies calculations in order to reduce the size of the model while maintaining data recognition accuracy.
- the analysis server 1200 transforms a convolutional filter in a convolutional neural network (CNN) to reduce the computational dimension or delete weights of the neural network that do not have a significant impact for data recognition. It performs pruning and quantization processes that simplify calculations by reducing the floating point number of weight values, enabling data lightweighting.
- the analysis server 1200 simplifies computation and maintains accuracy by imitating the output of a pre-trained large neural network and learning it from a small neural network.
- the analysis server 1200 may include a control unit (not shown).
- the control unit (not shown) can control the operation of each module included in the analysis server 1200, and can also control the operation of the smart watch 1100 (for example, displaying the screen of the smart watch).
- the analysis server 1200 can classify and predict the user's current sleep stage using sleep biometric data such as the user's heart rate and movement based on a previously learned deep learning model. there is.
- the analysis server 1200 can provide a wake-up alarm to help the user wake up in a light sleep stage, thereby minimizing the user's fatigue and helping the user wake up refreshed.
- the analysis server 1200 can find the optimal wake-up time and bedtime according to the user's sleep state through a pre-learned deep learning model and provide a wake-up alarm that allows the user to wake up accordingly, especially for Apple watch OS and Google wear. By implementing it as an OS application, a wake-up alarm can be provided through the smart watch (1100).
- the previously learned deep learning model used in the analysis server 1200 to classify and predict the user's sleep stage may be differently referred to as an AI algorithm or the like in the present invention.
- This pre-trained deep learning model (AI algorithm) is capable of equally classifying whether the user is currently in the REM sleep stage with a 100% probability when compared to the results of polysomnography diagnosed at existing hospitals. It can be provided.
- the analysis server 1200 can classify and predict the user's sleep stage by learning the user's biometric data collected through the smart watch 1100 using a deep learning model.
- the prediction module 1220 in the analysis server 1200 uses, for example, a plurality of smart watches owned by a plurality of users in the data collection module 1210 to implement (generate) a deep learning model for predicting sleep stages.
- a plurality of biometric data stripe biometric data and biometric data containing biometric information
- preprocessing is performed on the plurality of collected biometric data
- ii) the plurality of preprocessed biometric data is processed.
- Data merging is performed on the data, and then iii) the merged data can be applied as input to the deep learning model to train the deep learning model to classify sleep stages.
- FIGS. 10 to 14 The explanation for this can be more easily understood by referring to FIGS. 10 to 14.
- 10 to 14 are diagrams for explaining a deep learning model used in an analysis server according to an embodiment of the present invention.
- the prediction module 1220 monitors the data (biometric data) such as heart rate, number of steps, and movement collected for each user at different times and the sampling rate of the data. ) are all different, so considering this, the data is composed by cross-joining multiple collected biometric data according to time, and the data composed by cross-joining is then scaled to the user through a scaler. By eliminating the differences between the data and removing imbalanced data, multiple preprocessed biometric data can be prepared. For example, an example of data preprocessed by the prediction module 1220 (a plurality of preprocessed biometric data) may be as shown in FIG. 10.
- the scaler uses the original data as is when training a deep learning model, learning often slows down or problems occur as the original data has unique characteristics and distribution, so these problems arise. It may refer to a process of normalizing the collected original data (i.e., a plurality of collected biometric data) by scaling them equally to a certain range.
- the prediction module 1220 After performing preprocessing, when performing data merge, the prediction module 1220 combines (merges) all the preprocessed biometric data (i.e., biometric data of all multiple users) for which the scaler has been performed, and then labels them. Imbalanced data processing can be performed according to (Label). As an example, the relationship graph between merged data and labels and heart rate may be as shown in FIG. 11.
- imbalanced data means that one class has more numbers than the other class.
- the prediction module 1220 uses a method for evaluating the model using an evaluation index for class imbalance (e.g., ROC-AUC evaluation index), a tree-based machine learning algorithm that processes imbalanced data well (e.g., a doctor's Decision trees, random forests, etc.), resampling methods, etc. can be used, but the method is not limited to this, and various imbalanced data processing techniques known in the past or developed in the future can be applied. .
- an evaluation index for class imbalance e.g., ROC-AUC evaluation index
- a tree-based machine learning algorithm that processes imbalanced data well (e.g., a doctor's Decision trees, random forests, etc.), resampling methods, etc.
- resampling methods etc.
- the prediction module 1220 can apply the merged data as an input to the deep learning model to train the deep learning model to classify the sleep stage.
- a tree-based method such as Xgboost Sleep stage classification can be performed on data merged through machine learning, and 2D classification (i.e., data classification considering time and order) can be performed through deep learning neural networks.
- the error matrix (Confusion matrix) used by the prediction module 1220 when classifying sleep stages using machine learning (machine learning) may be, for example, as shown in FIG. 12, and the classification result may be as shown in FIG. 14. You can.
- the prediction module 1220 builds a previously learned deep learning model that can classify the REM sleep stage with 100% probability through biometric data such as movement and heart rate, and classifies and classifies the user's current sleep stage through this. It is predictable. Additionally, the analysis server 1200 may provide the following functions.
- the weather condition confirmation module 1240 in the analysis server 1200 causes the smart watch 1100 to provide a weather alarm as mentioned above, and then checks whether the user is in a weather condition, and at this time, determines whether the user is in a weather condition. If it is confirmed that the user's biometric data is continuously collected (obtained) through the smart watch (1100), but if it is confirmed that it is not a weather condition, the weather alarm is repeatedly issued by the smart watch (1100) through the re-alarm generation module (1250). It can be provided again.
- the waking state confirmation module 1240 sets exercise after waking up as the target action to reduce the sleep inertia of the user with reduced cognitive function.
- a wake-up alarm can be repeatedly provided to the user through the smart watch 1100 at preset intervals (for example, 5 minutes) until the set target behavior is recognized by the smart watch 1100.
- the weather condition confirmation module 1240 checks the preset target behavior and then performs the confirmed behavior in the smart watch 1100 based on the preset target behavior.
- a wake-up alarm can be generated and provided repeatedly until biometric data corresponding to the target behavior is obtained (that is, until it is detected that the user has taken the target behavior as a result of analysis of biometric data collected in real time).
- the weather condition confirmation module 1240 can repeatedly (continuously) provide a wake-up alarm at preset intervals until biometric data corresponding to the target behavior is acquired.
- the provision of the wake-up alarm is terminated so that the wake-up alarm is no longer provided to the smart watch 1100. (i.e., the provision of wake-up alarms can be turned OFF).
- the target action refers to exercise information that the user must take after waking up, and may be referred to differently by terms such as exercise after waking up or exercise mission information after waking up.
- the goal action may mean mission information that must be performed to turn off the wake-up alarm provided by the user.
- target actions include, for example, walking 10 steps (10 steps) after waking up, walking for 3 minutes, shooting a QR/barcode, performing NFC tagging between the smart watch (1100) and a user terminal (e.g., a smartphone, etc.), and walking for 3 minutes.
- This may be waving your hand, taking a picture of a cup, doing 10 squats, brushing your teeth after waking up, or jumping rope after waking up, but it is not limited to these, and various exercises and actions can be set as target actions.
- This goal behavior may be set automatically by the analysis server 1200, or may be set by receiving input from the user.
- the analysis server 1200 calculates the wake-up time according to the optimal sleep cycle and then transmits the information on the calculated wake-up time to the smart watch 1100, so that the smart watch 1100 calculates the wake-up time.
- the operation of the smart watch 1100 can be controlled to provide a wake-up alarm to the user at a time corresponding to the time.
- the analysis server 1200 detects that the power of the smart watch 1100 is in an OFF state (discharged state) at the calculated wake-up time, the analysis server 1200 sends the user terminal (not shown) holding information on the calculated wake-up time to the user. ), the operation of the user terminal can be controlled so that a wake-up alarm (wake-up alarm) is sounded and provided to the user at a time corresponding to the wake-up time calculated by the user terminal.
- the analysis server 1200 can provide a backup alarm function that causes a wake-up alarm to sound in the user terminal at the alarm setting time (i.e., calculated wake-up time) when the power is discharged while the smart watch 1100 is in use. .
- the analysis server 1200 controls the smart watch 1100 to provide a wake-up alarm at the wake-up time, but checks the power of the smart watch at the wake-up time to determine whether the smart watch is in an OFF state (i.e., in a discharge state). If it is detected, the power ON/OFF status of at least one user terminal owned by the user is checked to provide a backup alarm function, and at least one user terminal owned by the user whose power is in the ON state ( That is, it is possible to control the wake-up alarm to be provided from the user terminal (confirmed to be powered on).
- at least one user terminal possessed by the user may include various types of portable terminals, such as a smartphone, a tablet PC, and another smart watch.
- This function may be referred to as a backup alarm function in the present invention.
- the analysis server 1200 provides this backup alarm function, so that a specific user terminal that has received wake-up time information among a plurality of user terminals (e.g., smart watch, smartphone, tablet PC, etc.) owned by the user is in a discharged state. Accordingly, when it is impossible to provide a wake-up alarm, the wake-up alarm can be controlled to sound in a user terminal that is not in a discharged state (i.e., a user terminal that is turned on) in a complementary manner with other user terminals. there is.
- a discharged state i.e., a user terminal that is turned on
- the analysis server 1200 provides a wake-up alarm to the user, for example, the first step of outputting a quiet natural sound that soothes the user's mind through the smart watch 1100 is to the smart watch 1100.
- a wake-up alarm can be provided in the following order: step 2 of causing vibration, step 3 of outputting an alarm ringtone from the smart watch 1100, and step 4 of outputting an alarm ringtone from the user terminal.
- the analysis server 1200 when the analysis server 1200 is in a state where a wake-up alarm is being provided through the smart watch 1100 (i.e., when the wake-up alarm is ringing in the smart watch 1100), for example, the user terminal is connected to the smart watch.
- a weather alarm is provided only when the user terminal is detected to be held in the user's hand while being located within a short-distance communication distance (e.g., NFC, Bluetooth communication distance) that can be linked (connected) with (1100). This can be controlled to turn off (i.e., turn off the provided wake-up alarm).
- the analysis server 1200 is a remote (near-distance) location where the user can link his/her user terminal (e.g.
- the smart watch 1100 when a weather alarm is being provided. It can be controlled so that the wake-up alarm is turned off only when the user reaches the user terminal.
- the analysis server 1200 detects that the user wakes up, for example, through the AI algorithm of the smart watch 1100, or through the analysis of biometric data collected in real time by the analysis server 1200. If it detects that you are in a weather state, you can control the wake-up alarm to be automatically canceled (stopped, turned off).
- the analysis server 1200 considers the user's sleep stage (current sleep stage) analyzed using biometric information received in real time from the smart watch 1100 and generates a plurality of brain wave-induced sounds to induce sleep of the user. At least one brain wave-induced sound may be provided to the user terminal or smart watch 1100. Specifically, the analysis server 1200 provides optimized sleep based on information about the user's sleep stage (current sleep stage) analyzed (identified) with the user's biometric data collected from the smart watch 1100. To induce sleep, among a plurality of preset brain wave induction sounds for each sleep stage, the brain wave induction sound corresponding to the identified user's sleep stage can be controlled to be played through the user terminal or smart watch 1100.
- the analysis server 1200 monitors the user's sleep stage in real time to induce optimized sleep for the user, and generates brain wave-induced sounds suitable for each sleep stage according to changes in the sleep stage through the user terminal or smart watch ( 1100) can be played automatically.
- the analysis server 1200 automatically selects a sound source for an appropriate EEG-induced sound according to the user's status (e.g., which sleep stage the user is currently in) determined based on the user's biometric data. , and automatically adjust the volume so that the selected sound source is played on the user terminal or smart watch 1100.
- the brain wave-induced sound is a sound that induces sleep (sleep-inducing sound), and the analysis server 1200 can induce a good sleep in the user by playing the brain wave-induced sound according to the user's sleep stage (stage of the sleep cycle). .
- the plurality of EEG-induced sounds preset for each sleep stage are: i) the first EEG-induced sound played when the sleep stage is 'awake or almost awake (i.e., pre-sleep stage or waking state after REM sleep stage)'; , ii) the 2nd EEG-induced sound played when the sleep stage is 'stage 1 sleep stage', iii) the 3rd EEG-induced sound played when the sleep stage is 'stage 2 sleep stage', iv) the 3rd EEG-induced sound played when the sleep stage is 'stage 3'
- the 4th EEG-induced sound played when the sleep stage is 'Stage 4 sleep stage', v) The 5th EEG-induced sound played when the sleep stage is 'Stage 4 sleep stage', and vi)
- the sleep stage is 'REM sleep stage' It may include a sixth brain wave-induced sound that is played when.
- the first brain wave-induced sound may mean an 8Hz sound that induces low alpha waves
- the second brain wave-induced sound may mean a 5Hz sound that induces medium theta waves
- the fourth brain wave-induced sound is a sound that induces delta waves, and may mean a sound having any one hertz (Hz) value between 1Hz and 3Hz
- the sixth brain wave-induced sound may include at least one of a 4Hz sound that induces low theta waves and a sound that induces gamma waves (i.e., a sound having any one hertz value between 30Hz and 45Hz).
- the analysis server 1200 provides an 8Hz sound (i.e., a first brain wave-induced sound) that induces low alpha waves when the user is determined to be in a state before going to bed or waking up. ) can be played automatically.
- 8Hz sound i.e., a first brain wave-induced sound
- low alpha waves mimic natural brain wave frequencies before going to bed and just before waking up
- the analysis server 1200 provides a first brain wave inducing sound that induces low alpha waves, allowing the user to gently fall asleep and wake up from sleep. can help you do it.
- the analysis server 1200 automatically plays a 5Hz sound (i.e., a second brain wave-induced sound) that induces intermediate theta waves on the smartphone or smart watch 1100 when it is determined that the user is in stage 1 sleep. can do.
- a 5Hz sound i.e., a second brain wave-induced sound
- the middle theta waves can help users sleep comfortably by relieving tension.
- a 1 to 3 Hz sound i.e., fourth brain wave inducing sound
- a 1 to 3 Hz sound i.e., fourth brain wave inducing sound
- Delta waves can help you fall into the deepest sleep of your sleep cycle.
- the analysis server 1200 automatically plays the 4Hz sound that induces low theta waves and the 30-45Hz sound that induces gamma waves on the smartphone or smart watch (1100) when it is determined that the user is in the REM sleep stage. can do.
- the analysis server 1200 is illustrated as providing brain wave-induced sound (i.e., sound, sound source, sound) to induce sleep in the user, but the present invention is not limited thereto, and the analysis server 1200 provides sleep inducing sound to the user.
- Various types of content such as guided videos, can also be provided.
- the analysis server 1200 of the present invention can provide sleep coaching in real time through methods such as light therapy and brain wave induction by monitoring the user's sleep stage in real time using the smart watch 1100.
- the analysis server 1200 can control ON/OFF of the blue light blocking mode and ON/OFF of the blue light enhancement mode for the smart watch 1100 in consideration of the calculated wake-up time information.
- the blue light enhancement mode considered in the present invention unlike the ON/OFF function of the blue light blocking mode, makes the display screen of the user terminal blue overall when the user uses the user terminal (for example, a smartphone, etc.) in the morning. It can mean a function that does. A more specific explanation for this is as follows.
- FIG. 14 is a diagram for explaining the light spectrum of a smartphone.
- melatonin is a sleep hormone that is secreted in small amounts during the day and in large quantities at night, thereby controlling and inducing sleep and acting on the circadian rhythm.
- melatonin is suppressed during the day and produced at night to promote sound sleep.
- our body recognizes light in the 400 nm spectrum, which is the wavelength of sunlight, and suppresses melatonin.
- blue light refers to blue visible light with a wavelength of 380 nm emitted from smartphones, etc. This blue light has the characteristic of disrupting the user's biological clock rhythm by making the user think it is daytime even at night.
- the analysis server 1200 calculates the optimal bedtime and optimal wake-up time that optimize sleep efficiency, sets the calculated optimal bedtime as the user's bedtime alarm time, and sets the calculated optimal wake-up time to the user's bedtime. You can set the wake-up alarm time.
- the analysis server 1200 ensures that a sleep alarm is provided through the smart watch 1100 at a time corresponding to the optimal bedtime (i.e., bedtime alarm time), and at a time corresponding to the optimal wake-up time (i.e., wake-up alarm time).
- a wake-up alarm can be provided through the smart watch 1100. At this time, let us assume that the optimal bedtime is 11 PM and the optimal wake-up time is calculated to be 4 AM.
- the analysis server 1200 uses a smart watch during the night time section corresponding to before the optimal bedtime (for example, before a preset time from the optimal bedtime) to before the optimal wake-up time (for example, from 8 PM to 4 AM).
- (1100) and the user terminal can each be controlled to operate in the blue light blocking mode (i.e., control the blue light blocking mode to ON).
- the analysis server 1200 turns on the blue light enhancement mode for each of the smart watch 1100 and the user terminal (not shown). It can be controlled with .
- morning may illustratively mean some of the time sections (for example, the daytime section between 4:00 AM and 8:00 PM) excluding the night time section described above.
- the analysis server 1200 can turn on the blue light blocking mode and turn off the blue light enhancement mode for the smart watch 1100 and the user terminal (not shown) during the night time section, and turn off the blue light enhancement mode during the day. In addition to turning off the blue light blocking mode, you can also turn on the blue light enhancement mode.
- the analysis server 1200 provides this blue light enhancement mode function, so that when a user uses a user terminal (smart phone, mobile phone) or smart watch in the morning (particularly during the above-mentioned daytime period), a blue-based color is displayed on the display.
- the included light is controlled in a mode that emits light (i.e., a mode that emits light with a wavelength belonging to the first wavelength range, which will be described later), providing the effect of awakening the user in the morning (morning), thereby allowing the user to You can set your desired circadian rhythm. This is because in the case of phototherapy (light therapy), exposure to blue light for at least 30 minutes and less than 1 hour is sufficient, regardless of the intensity of the light.
- the blue light enhancement mode is to provide an awakening effect to the user and suppress melatonin secretion, corresponding to 380 nm to 500 nm on the screen (LCD screen) and LED display unit of the smart watch 1100 and the user terminal, respectively.
- This may mean a mode in which light of any one wavelength in the first wavelength range is provided (emitted) (i.e., a mode in which light containing a blue color is emitted).
- the blue light blocking mode in order to provide the user with a good sleep effect and produce (secrete) melatonin, the screen (LCD screen) and LED display unit of the smart watch (1100) and the user terminal respectively exceed 500 nm and below 700 nm. It may refer to a mode in which light of any one wavelength in the corresponding second wavelength range is provided (emitted) (i.e., a mode in which light containing an orange-based color is emitted).
- the analysis server 1200 controls the ON/OFF of the blue light blocking mode or blue light enhancement mode for the smart watch 1100 and the user terminal owned by the user, but only in this case. It is not limited, and the analysis server 1200 can control the ON/OFF of the blue light blocking mode for the display (screen) of various electronic devices (for example, VR devices, computer monitors, TVs, etc.) owned by the user. It may be possible. To this end, for example, the user may pre-register electronic devices for which they wish to control the ON/OFF of the blue light blocking mode or blue light enhancement mode in the analysis server 1200 through the user terminal.
- the analysis server 1200 provides an ON/OFF control function of the blue light blocking mode or blue light enhancement mode, thereby providing an awakening effect by emitting light with a spectrum wavelength belonging to the first wavelength range as blue light during the daytime section.
- a light therapy function that provides a sound sleep effect by emitting light of a spectrum wavelength belonging to the second wavelength range during the night time period can be implemented.
- the analysis server 1200 may provide a blue light blocking mode.
- the principle of blocking blue light, which disrupts the biological clock rhythm by making people think it is daytime even at night, with a program is to reduce the blue light source among the RGB (red, green, blue) colors that make up the digital screen.
- the display In the existing blue light blocking mode, the display (screen) darkens and changes to warm colors, as if applying night mode (dark mode). In addition, depending on the existing program, it is possible to set the blue light to be blocked to the extent desired by the user (i.e., blue light blocking intensity) and at the desired time (a specific time designated directly).
- the existing night mode (dark mode) function has the characteristic of causing myopia and astigmatism.
- the existing night mode (dark mode) provides a user environment (UI) where bright text appears on a dark background.
- UI user environment
- the pupil expands and the light entering the pupil is concentrated in one place.
- the position of the lens inside the eye tends to move forward, causing myopia.
- the present invention improves the disadvantages of the existing blue light blocking mode setting method and the conventional night mode (dark mode) described above, and automatically blocks blue light when the time appropriate for the user's individual bedtime is reached through an algorithm.
- the blue light blocking mode is controlled to ON, and when the time corresponding to the optimal wake-up time arrives (arrival), the blue light blocking mode is automatically turned OFF for the smart watch (1100) and the user terminal, thereby turning the blue light mode on. It can be controlled to be ON.
- the time setting for ON/OFF control of blue light was set to a time directly specified by the user (i.e., manually specified), whereas in the present invention, the analysis server 1200 automatically calculates the time (i.e. It can be set automatically based on optimal sleep time and optimal wake-up time.
- the analysis server 1200 automatically displays a spectrum wavelength of 480 nm on the screen (display) of the smart watch 1100 and the user terminal when the time for the user to wake up arrives (i.e., just before the optimal wake-up time). It can provide an awakening effect by emitting (providing) blue light. Thereafter, the analysis server 1200 may set the blue light enhancement mode to ON during morning hours. Thereafter, when the user's bedtime arrives (i.e., just before the optimal bedtime), the analysis server 1200 automatically displays, for example, a spectrum of 580 nm on the screen (display) of the smart watch 1100 and the user terminal. It can provide a sound sleep effect by emitting (providing) orange light of a certain wavelength.
- the analysis server 1200 can attach a blue light blocking film to the flash light LED on the rear of the user terminal and control the rear flash light LED to emit light when using the mobile phone at night, as will be described later.
- the analysis server 1200 enhances the blue light of the display with a program when the user uses the user terminal (smart phone, mobile phone) in the morning (especially during the above-mentioned daytime section).
- enhanced blue light may mean that the blue light blocking mode is controlled to OFF), providing the effect of awakening the user in the morning (morning) so that the user can set the desired circadian rhythm. can do. This is because in the case of phototherapy (light therapy), exposure to blue light for at least 30 minutes and less than 1 hour is sufficient, regardless of the intensity of the light. On the other hand, depending on the intensity of white light, it is recommended to use 10,000 lux for 30 minutes and 2,500 lux for 2 hours.
- the analysis server 1200 may provide a blue light blocking flash light LED for blocking blue light.
- the best way to use a cell phone at night while protecting the user's eyes is to not look at the cell phone at night with the lights off.
- viewing the original screen of a cell phone in a bright place with the lights turned on is the best way to avoid damaging one's eyes.
- using a cell phone in a bright place with the lights on like this has the problem of disrupting the user's sleep due to the blue light from the cell phone and lights.
- the analysis server 1200 attaches a blue light blocking film to the flash light LED on the back of the user terminal carried by the user, and then at night (particularly during the night time described above). (during the section), when it is detected that the user is using the user terminal, the flash light LED on the back of the user terminal is controlled to emit light, thereby providing light with blue light blocked.
- the analysis server 1200 may install a self-developed blue light blocking flash light LED provided by the analysis server 1200 on the rear of the user terminal.
- the self-developed blue light blocking flash light LED may, for example, be provided in a form in which a blue light blocking film is attached to the basic flash light LED, and may be installed in a user terminal.
- the blue light blocking flash light LED may, for example, be prepared to emit (irradiate, provide) light of a wavelength belonging to a second wavelength range (i.e., a wavelength range of more than 500 nm and less than 700 nm).
- This analysis server 1200 provides a blue light blocking flash light LED, allowing the user to use the mobile phone while protecting the eyes even when use of the mobile phone is unavoidable at night.
- the analysis server 1200 provides a blue light blocking mode ON/OFF control function, a blue light enhancement mode function (i.e., a blue light enhancement ON/OFF function), and a blue light blocking flash light LED, thereby providing a user terminal ( In other words, it is possible to effectively prevent the user from experiencing myopia and astigmatism due to use of a mobile phone or smart watch 1100.
- a blue light blocking mode ON/OFF control function i.e., a blue light enhancement ON/OFF function
- a blue light blocking flash light LED i.e., a blue light enhancement ON/OFF function
- the analysis server 1200 provides a blue light blocking function (in particular, a blue light blocking function by controlling the display screen or installing a blue light blocking flash light LED on the back of the user terminal). can do.
- a blue light blocking function in particular, a blue light blocking function by controlling the display screen or installing a blue light blocking flash light LED on the back of the user terminal.
- the analysis server 1200 is provided with a 500-700nm wavelength (i.e., a second wavelength that blocks blue light) on the back and sides of various display devices (e.g., smartphones, TVs, monitors, smart watches, etc.) considered in the present invention.
- a background LED in other words, background light LED
- the analysis server 1200 installs hardware or detachable LEDs on the back and sides of the display device to prevent myopia and astigmatism that may occur when the user uses the display device at night (this blocks the background LED or blue light described above). Flash light (meaning LED) can be installed.
- the display device refers to various terminal devices considered in the present invention and may include devices such as user terminals (eg, smartphones, etc.), smart watches, TVs, and monitors.
- the analysis server 1200 can provide the user with a background light LED that can be installed in one area (for example, the back and sides) of the display device, including the smart watch and the user terminal carried by the user, where the background light LEDs can be provided in a form that can prevent myopia and astigmatism due to use of the display device when a user must use the display device at night.
- the background light LED can be installed as hardware to the display device or can be installed in a detachable manner.
- the analysis server 1200 of the present invention can prevent myopia, astigmatism, and sleep disorders that may occur when using a display device at night through display application software and hardware.
- the background light LED i.e., background LED, blue light blocking flash light LED
- the background light LED i.e., background LED, blue light blocking flash light LED
- the background light LED installed in the display device i.e., background LED, blue light blocking flash light LED
- the flash light LED provided on the rear of the display device i.e., basic flash light LED, which is a typical light
- It may be arranged to enable use, setting, and operation by control by the analysis server 1200 or control based on user input.
- the background light LED provided to prevent myopia and astigmatism and the flash light LED (ordinary light) on the back of the display device are controlled by human settings of software such as an application that operates them. They can be arranged to be controlled, and they can also be arranged to be usable not only in the real-time sleep coaching technology proposed by the present invention but also in other general applications.
- the background light LED may be provided so that the background light LED capable of preventing myopia and astigmatism can be installed not only in the display device pre-registered by the user but also when manufacturing the display device.
- the display device pre-registered by the user includes various electronic devices owned by the user, including smart watches and user terminals (smartphones, etc.) (for example, various types of display devices such as VR devices, computer monitors, TVs, etc.) may be included.
- the present system 11 i.e., a real-time sleep health management service providing system using AI-based brain wave tuning and autonomic nervous system control
- the present invention is capable of preventing myopia and astigmatism even when a user uses a display device in a normal situation. Since it is possible to install a background light LED, the occurrence of myopia and astigmatism due to the use of the display device can be effectively prevented (prevented).
- the existing blue light blocking mode darkens the display when turned ON, blocking blue light of 400 to 500 nm and converting it to a warmer color in the color spectrum. However, this also darkens the display, which has the disadvantage of causing myopia and astigmatism. there is.
- the analysis server 1200 may turn on the blue light blocking mode while the display device is brightened (that is, the display screen has a brightness higher than a preset brightness value).
- the analysis server 1200 measures the amount of ambient light of the display device using an illumination sensor (light sensor) of the display device, and operates the light emission of the above-described background LED based on the measured amount of ambient light. You can also control it. That is, the analysis server 1200 can measure the amount of surrounding light using the illuminance sensor (light sensor) of the display device and link the background LED emission.
- an illumination sensor light sensor
- the analysis server 1200 can measure the amount of surrounding light using the illuminance sensor (light sensor) of the display device and link the background LED emission.
- the analysis server 1200 can derive the user's optimal sleep time, nap time, and bedtime (i.e., bedtime) based on analysis of biometric data, and provide an alarm function based on this.
- the analysis server 1200 can provide hypnosis and breathing-related images related to sleep techniques (for example, techniques such as nervous fatigue sleep mode and breathing training) to the smart watch 1100 or the user terminal.
- sleep techniques for example, techniques such as nervous fatigue sleep mode and breathing training
- the analysis server 1200 can provide hypnosis and breathing-related images related to sleep techniques (for example, techniques such as nervous fatigue sleep mode and breathing training) to the smart watch 1100 or the user terminal.
- sleep techniques for example, techniques such as nervous fatigue sleep mode and breathing training
- the breathing-related image is converted into an image based on sunset light (orange color) corresponding to light with a wavelength of 550 to 600 nm. It is possible to reproduce and provide (simultaneously) the regenerated breathing-related images to the smart watch 1100 as well as the user terminal, thereby allowing the smart watch and the user terminal to play the regenerated breathing-related images.
- the analysis server 1200 may provide breathing training-related videos (e.g., video content related to breathing training, such as breathing in through the nose) to the smart watch 1100.
- breathing training-related videos e.g., video content related to breathing training, such as breathing in through the nose
- the breathing training-related video also, an image based on sunset light (orange color) corresponding to light with a wavelength of 550 to 600 nm can be reproduced and provided to the smart watch (1100).
- the analysis server 1200 calculates the wake-up time using an artificial intelligence model for real-time sleep coaching for the user, provides a wake-up alarm based on the calculated wake-up time, and provides a blue light blocking mode.
- Various functions can be provided to the user, such as ON/OFF control function, ON/OFF control function of blue light augmentation mode, function to provide background light LED, function to provide brain wave-induced sound, and backup alarm function.
- the analysis server 1200 is not prepared to provide all of these various functions based on an artificial intelligence model, but rather provides at least some of the various functions (for example, blue light Augmented mode ON/OFF control function, backup alarm function, etc.) do not need to be limited to artificial intelligence and can be provided to users without an artificial intelligence model. That is, at least some of the various functions provided by the analysis server 1200 (for example, the ON/OFF control function of the blue light augmentation mode, the backup alarm function, etc.) are not limited to artificial intelligence, but Even without an artificial intelligence model, it can be applied (implemented, prepared) for general use by expanding its scope so that it can be used by setting by a human (user) or in general situations.
- the various functions for example, blue light Augmented mode ON/OFF control function, backup alarm function, etc.
- Figure 15 is a diagram showing the signal flow of a real-time sleep health management service providing system using AI-based brain wave entrainment and autonomic nervous system control according to an embodiment of the present invention.
- Figure 15 is a diagram illustrating a schematic operational flow of a method for providing real-time sleep health management service using AI-based brain wave tuning and autonomic nervous system control according to an embodiment of the present invention.
- the method of providing real-time sleep health management service using AI-based brain wave entrainment and autonomic nervous system control shown in FIG. 15 can be performed by the system 11 described above. Therefore, even if the content is omitted below, the content described about the system 11 can be equally applied to the explanation of the method of providing real-time sleep health management service using AI-based brain wave entrainment and autonomic nervous system control.
- step S150 the smart watch 1100 collects sleep biometric data including the user's heart rate, movement, and blood pressure while sleeping, and transmits the collected sleep biometric data to the analysis server 1200.
- step S155 the analysis server 1200 analyzes sleep biometric data, calculates optimal sleep cycle information and wake-up time in step S160, and transmits the calculated optimal sleep cycle information and wake-up time information to the smart watch 1100.
- step S165 a wake-up alarm (wake-up alarm) is provided to the user according to the received wake-up time, and in step S170, the user's biometric information is collected.
- the user biometric information is biometric information during the user's activities and may include heart rate, blood pressure, and movement information.
- step S175 the user's biometric information is collected from the smart watch 1100 to check whether the user is awake (i.e., whether the user is awake). If it is confirmed that the user has woken up (i.e., is confirmed to be in a waking state) in step S175, step S150 is re-entered, and if it is determined that the user is not woken up in step S175, step S180 is entered. In step S180, the smart watch 1100 provides a re-alarm to help the user wake up.
- step S180 after providing a re-alarm in step S180, it is possible to re-check whether the user is awake (i.e., whether he is awake) by entering step S175 again, and repeatedly providing an alarm until the user fully wakes up (i.e. Step S180 can be repeated). Additionally, at this time, if the user touches the wake-up alarm end icon on the smart watch 1100, the alarm repetition can end.
- steps S150 to S180 may be further divided into additional steps or combined into fewer steps, depending on the implementation of the present invention. Additionally, some steps may be omitted or the order between steps may be changed as needed.
- Figure 16 is a diagram showing a method of providing a wake-up alarm according to an embodiment of the present invention.
- the method of providing a weather alarm shown in FIG. 16 may be performed by the analysis server 1200 described above. Therefore, even if the content is omitted below, the content described with respect to the analysis server 1200 can be equally applied to the explanation of the method of providing a weather alarm.
- step S161 the analysis server uses a smart watch to classify the presence (or absence) of the user's movement.
- heart rate data is accumulated when the user does not move, and a deep learning data model is learned to predict the user's sleep stage.
- step S163 the user's sleep stage and heart rate change are considered to determine whether the user is in the REM sleep stage, and in step S164, a wake-up alarm (wake-up notification) is sent through the smart watch (1100) in the light sleep stage after the REM sleep stage. shall be provided.
- the analysis server may repeatedly generate and provide an alarm until, for example, it confirms that the user's movement and heart rate are in a waking state (that is, a waking-up alarm may be repeatedly generated and provided through a smart watch).
- steps S161 to S165 may be further divided into additional steps or combined into fewer steps, depending on the implementation of the present invention. Additionally, some steps may be omitted or the order between steps may be changed as needed.
- Figure 17 is a diagram showing a method of adjusting the wake-up time reflecting the user's actual sleep time according to an embodiment of the present invention.
- the method of adjusting the wake-up time that reflects the user's actual sleep time shown in FIG. 17 can be performed by the analysis server 1200 described above. Therefore, even if the content is omitted below, the content described with respect to the analysis server 1200 can be equally applied to the explanation of the method of adjusting the wake-up time to reflect the user's actual sleep time.
- step S1100 the analysis server collects biometric information from the smart watch, and in step S1200, the sleep start time when the user goes to sleep is determined through biometric information analysis.
- step S1300 the sleep start time is compared with the usual pattern.
- step S1400 the wake-up time is adjusted by considering the repeated sleep cycle according to the comparison results.
- step S1400 the analysis server determines that the user is in the light sleep stage of the last sleep cycle predicted by considering the remaining time to sleep until the time to wake up.
- the wake-up time can be calculated so that the wake-up time can occur, and then the calculated wake-up time can be transmitted to the smart watch so that the user can receive a wake-up alarm at that time.
- the analysis server 1200 not only collects and analyzes biometric information such as the user's movements and heart rate using a smart watch, but also acquires sleep habit data for each user through deep learning.
- the analysis server 1200 calculates the difference from the acquired sleep habit data by comparing it with the sleep start time of the actual user who goes to sleep on average, and calculates a time equal to the calculated difference to enter the smart sleep stage in the last sleep cycle. You can wake the user through the watch. Specifically, if the user falls asleep 1 hour later or earlier than usual, the analysis server 1200 does not wake up at a set time, but only sleeps until the time after the 1.5-hour sleep cycle is repeated a certain number of times according to the sleep cycle graph. You can calculate the wake-up time so you can wake up. For example, the analysis server 1200 may calculate an alarm time so that the user wakes up after 4.5 hours when the sleep cycle is repeated 3 times or 6 hours after the sleep cycle is repeated 4 times.
- the analysis server 1200 calculates bedtime and wake-up time at a certain level or higher in sleep efficiency, which is the actual sleeping time compared to the time lying in bed, through sleep biometric data learned through machine learning, and the calculated bedtime and wake-up time. Be notified.
- the analysis server 1200 calculates and provides (notifies) the insufficient sleep time compared to the usual pattern and guides the user to supplement the insufficient sleep time with a nap or nap.
- Supplementary sleep time guidance information can be created and provided. For example, if the required sleep time per day is 8 hours, but the user can only sleep 5 hours, the analysis server 1200 can guide the user to make up for the 3 hours of sleep time by taking a nap or napping during the day. there is.
- the analysis server 1200 uses a smart watch to accumulate and analyze the user's sleep biometric data even when the user takes a nap, and wakes the user in consideration of the sleep cycle. Additionally, even in the case of a nap, the analysis server 1200 determines whether the user has completely woken up through biometric data analysis and provides a repetitive wake-up alarm. In addition, when the user usually takes a nap, the sleep time changes if the user falls asleep earlier or later than the usual sleep time, so the analysis server 1200 determines the light sleep during the last sleep cycle predicted by considering the time remaining until the time to wake up. The wake-up time can be calculated to wake up at the right time.
- steps S1100 to S1400 may be further divided into additional steps or combined into fewer steps, depending on the implementation of the present invention. Additionally, some steps may be omitted or the order between steps may be changed as needed.
- the present invention provides a real-time sleep health management service provision system (11) and method using AI-based brain wave entrainment and autonomic nervous system control, thereby providing individual sleep time, bedtime, wake-up time, and nap time.
- Customized management can be done according to the sleep cycle, and the sleep cycle can be used to ensure that the user can sleep refreshed even if he or she sleeps briefly. Through this, the quality of life is improved by increasing the user's time utility, sleep efficiency, and rest efficiency. Make it possible to do it.
- the present invention provides a real-time sleep health management service provision system (11) and method using AI-based brain wave entrainment and autonomic nervous system control, thereby reducing the required REM sleep time when the user does not recover sufficiently after sleep. It can inform the user or provide various services for user recovery, thereby speeding up the user's recovery from stress and fatigue.
- the present invention provides a real-time sleep health management service provision system (11) and method using AI-based brain wave entrainment and autonomic nervous system control, providing deep sleep based on a customized training data set optimized for sleep stage determination.
- a sleep stage judgment model can be implemented by performing machine learning based on a learning neural network, and based on this, the quality of sleep stage prediction and wake-up time calculation results calculated from the sleep stage judgment model can be further improved.
- the method of providing a real-time sleep health management service using AI-based brain wave entrainment and autonomic nervous system control includes commands executable by a computer, such as an application or program module executed by a computer. It can also be implemented in the form of a recording medium containing.
- Computer-readable media can be any available media that can be accessed by a computer and includes both volatile and non-volatile media, removable and non-removable media. Additionally, computer-readable media may include all computer storage media.
- Computer storage media includes both volatile and non-volatile, removable and non-removable media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules or other data.
- the method of providing a real-time sleep health management service using AI-based brain wave entrainment and autonomic nervous system control includes an application installed by default on a terminal (this is a program included in a platform or operating system, etc., which is basically installed on the terminal). may include), and may be executed by an application (i.e. program) installed directly on the master terminal by the user through an application providing server such as an application store server, an application, or a web server related to the service. .
- the method of providing a real-time sleep health management service using AI-based brain wave tuning and autonomic nervous system control is an application (i.e., program) installed by default on the terminal or directly installed by the user. It may be implemented and recorded on a computer-readable recording medium such as a terminal.
- the present invention uses light and sound to achieve brain wave synchronization, and uses light and breathing to control the autonomic nervous system, thereby adjusting the brain waves and autonomic nervous system to suit the bedtime and wake-up time, and to match the wake-up time.
- By playing a sound source of monaural beats or binaural beats that synchronize theta waves, alpha waves, and beta waves through a speaker it induces brain waves when waking up, and provides an awakening effect and serotonin by providing a blue light sunlight shower at the time of waking up.
- By converting the generated serotonin into melatonin at bedtime it is possible to further induce bedtime and sound sleep.
- alternating somatosensory stimulation is applied to the user's terminal to reduce stress. and wearable devices, and has the potential for industrial use as it can relieve cluster headaches caused by abnormal activity of the hypothalamus by normalizing the circadian rhythm and simultaneously relieve pain through the non-invasive vagus nerve.
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Animal Behavior & Ethology (AREA)
- Veterinary Medicine (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Physics & Mathematics (AREA)
- Psychology (AREA)
- Anesthesiology (AREA)
- Hematology (AREA)
- Primary Health Care (AREA)
- Pathology (AREA)
- Epidemiology (AREA)
- Acoustics & Sound (AREA)
- Pain & Pain Management (AREA)
- Molecular Biology (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Biophysics (AREA)
- Surgery (AREA)
- Child & Adolescent Psychology (AREA)
- Developmental Disabilities (AREA)
- Hospice & Palliative Care (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
Provided is a system for providing a real-time sleep health management service by using an AI-based brain wave synchronization and autonomic nervous system control, the system comprising: a user terminal for setting a bedtime and a wake-up time, emitting light at a melatonin non-suppressing wavelength while outputting a sleep induction method, which is a breathing technique that accelerates the parasympathetic nervous system at the bedtime, transmitting vibration through a wearable device linked to the user terminal and providing alternating somatosensory stimulation on both sides thereof so as to calm the sympathetic nervous system or accelerate the parasympathetic nervous system through vagus nerve stimulation, inserting and outputting a flicker of an alpha-band or a beta-band in order to synchronize alpha or beta brain waves at the wake-up time, and outputting blue light to induce wakening; and a management service provision server including a database unit for storing settings for light, sound, vibration and a breathing technique, which are for brain wave synchronization and autonomic nervous system control at the bedtime and wake-up time, a setting unit for receiving the bedtime and wake-up time settings from the user terminal, and a control unit for controlling that at least one from among the light, the sound, the vibration and the breathing technique for brain wave synchronization and autonomic nervous system control is output at the bedtime and wake-up time set in the user terminal.
Description
본 발명은 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템에 관한 것으로, 빛, 사운드, 호흡법 및 진동을 통해 뇌파동조 및 자율신경계조절을 수행함으로써 수면 및 일주기리듬을 정상화하는 시스템을 제공한다.The present invention relates to a real-time sleep health management service provision system using AI-based brain wave entrainment and autonomic nervous system regulation, which normalizes sleep and circadian rhythm by performing brain wave entrainment and autonomic nervous system regulation through light, sound, breathing, and vibration. Provides a system.
불면증과 같은 수면장애는 사회가 발전할수록 다양한 스트레스로 인해 발생한다. 한국표준 질병분류체계(Korean Standard Classification of Diseases, KCD)에서도 수면장애(Sleep Disorders)를 질병코드 G47로 분류하고 있다. G47의 소분류에는 불면증 관련 질병(G47.0)과 수면각성장애(G47.2)가 있다. 충분한 수면은 사람의 건강과 활력을 회복시킬 뿐만 아니라 여러 가지 인체의 호르몬과 밀접한 관련이 있다. 그래서 수면을 적절히 취하지 못할 경우 일상생활에서의 피로감 무기력감을 느낄 수 있지만 심할 경우에는 만성피로, 치매, 우울증, 공황장애와 같은 심각한 질병에 노출될 수 있다. 그렇기 때문에 현대인의 수면 장애율이 증가될수록 수면케어의 필요성도 높아지고 있다.Sleep disorders such as insomnia occur due to various stresses as society develops. The Korean Standard Classification of Diseases (KCD) also classifies sleep disorders with disease code G47. Subcategories of G47 include insomnia-related diseases (G47.0) and sleep-wake disorders (G47.2). Sufficient sleep not only restores a person's health and vitality, but is also closely related to various hormones in the human body. Therefore, if you do not get adequate sleep, you may feel fatigue and lethargy in daily life, but in severe cases, you may be exposed to serious diseases such as chronic fatigue, dementia, depression, and panic disorder. Therefore, as the rate of sleep disorders among modern people increases, the need for sleep care also increases.
이때, 모노럴비트로 뇌파동조를 일으켜 숙면을 돕거나 사운드 및 라이트 테라피를 이용하여 수면을 돕는 방법이 연구 및 개발되었는데, 이와 관련하여, 선행기술인 한국공개특허 제2023-0080260호(2023년06월07일 공개) 및 한국공개특허 제2012-0131253호(2012년12월05일 공개)에는, 파동형태의 모노럴비트를 믹싱하여 뇌파동조를 일으킬 수 있도록 하고, 뇌파동조를 위한 기준주파수에 따른 주파수데이터를 로딩하여 데시벨을 조정하고, 중첩된 모노럴비트를 웨이브파일로 생성하여 사용자 단말로 제공하는 구성과, 수면상태를 압전센서로 검출한 후 라이트 테라피 및 사운드 테라피를 제공하는 구성이 각각 개시되어 있다.At this time, a method of helping a good night's sleep by stimulating brain waves with monaural beats or helping sleep by using sound and light therapy was researched and developed. In this regard, the prior art, Korea Patent Publication No. 2023-0080260 (June 7, 2023), was researched and developed. published) and Korea Patent Publication No. 2012-0131253 (published on December 5, 2012), brain wave entrainment can be generated by mixing monaural beats in the form of waves, and frequency data according to the reference frequency for brain wave entrainment is loaded. A configuration that adjusts the decibel, generates overlapping monaural beats as a wave file and provides it to the user terminal, and a configuration that detects the sleep state with a piezoelectric sensor and then provides light therapy and sound therapy are disclosed, respectively.
다만, 전자의 경우 모노럴비트를 믹싱하는 방법만을 개시할 뿐 취침 및 기상에 따라 뇌파 및 자율신경계를 조절하는 구성은 개시되어 있지 않다. 후자의 경우에도 라이트 테라피 및 사운드 테라피로만 개시되어 있을 뿐 모노럴비트나 바이노럴비트를 이용하는 구성은 개시하고 있지 않으며, 라이트 테라피의 종류도 개시되어 있지 않다. 취침 및 기상에 따라 뇌파와 자율신경계가 조절되는데 불면증이나 과잉수면의 경우 뇌파와 자율신경계 조절이 정상범위에서 벗어난 경향을 보인다. 또, 일주기리듬을 담당하는 시상하부에 이상활성이 발생한 경우 일주기리듬이 깨질 뿐 아니라 군발두통(Cluster Headache)의 원인이 되고, 우울증, ADHD, 불안장애, 강박장애, 공황장애 등 다양한 증상에 이를 수 있다. 이에, 수면주기를 관리해줌으로써 궁극적으로 일주기리듬을 되찾도록 할 수 있는 시스템의 연구 및 개발이 요구된다.However, in the former case, only a method of mixing monaural beats is disclosed, and a configuration for controlling brain waves and autonomic nervous system according to sleeping and waking up is not disclosed. In the latter case, only light therapy and sound therapy are disclosed, and configurations using monaural beats or binaural beats are not disclosed, and the type of light therapy is not disclosed. Brain waves and autonomic nervous system are regulated depending on sleeping and waking up, but in the case of insomnia or excessive sleep, brain waves and autonomic nervous system regulation tend to deviate from the normal range. In addition, if abnormal activity occurs in the hypothalamus, which is responsible for the circadian rhythm, not only the circadian rhythm is broken, but it can also cause cluster headaches and cause various symptoms such as depression, ADHD, anxiety disorder, obsessive-compulsive disorder, and panic disorder. This can be achieved. Accordingly, research and development of a system that can ultimately restore circadian rhythm by managing the sleep cycle is required.
사람은 잠을 자면서 약 1시간 30분 정도의 주기로 얕은 수면과 깊은 수면, 얕은 수면, 꿈 수면(램 수면)의 과정을 반복한다. 일반적인 수면주기를 나타낸 도 6을 참조하면, 개인마다 차이는 있지만 보통 얕은 수면에서 꿈 수면까지 한 주기를 순회하는 시간은 약 1시간 30분 정도이다. 대부분의 사람들이 6~7시간을 잔다고 간주한다면 총 4~5번의 주기를 그리면서 수면을 취하는 것이다.While sleeping, people repeat the processes of light sleep, deep sleep, light sleep, and dream sleep (REM sleep) in a cycle of about 1 hour and 30 minutes. Referring to Figure 6, which shows a general sleep cycle, although it varies from person to person, the time it takes to go through one cycle from light sleep to dream sleep is usually about 1 hour and 30 minutes. If we assume that most people sleep for 6 to 7 hours, then they sleep in a total of 4 to 5 cycles.
조금 밖에 잠을 못 잤는데도 잠자리에서 일어날 때 어느 날은 개운하게 일어나는 반면, 어떤 날은 힘들게 일어났던 경험이 있을 것이다. 짧게 잠을 자더라도 개운하게 일어나려면 얕은 수면단계에서 깨는 것이 좋다. 그러므로 최소한의 수면시간과 수면리듬을 고려한다면 수면주기를 2~3바퀴 돌고 난 3시간에서 4시간 30분 정도의 수면 뒤에 기상하는 것이 효율적인 수면을 취한 느낌을 가질 수 있다. 예컨대, 1시에 잔다고 가정하면 5시 30분이나 40분쯤에 잠이 깨도록 알람을 설정하고 기상하면 짧지만 깊은 잠을 잔 듯한 개운함을 느낄 수 있다.Even though you only got a little sleep, you may have experienced waking up feeling refreshed on some days and having a hard time on other days. Even if you sleep briefly, it is best to wake up in a light sleep stage to wake up feeling refreshed. Therefore, considering the minimum sleep time and sleep rhythm, waking up after 3 to 4 hours and 30 minutes of sleep after completing 2 to 3 sleep cycles can give you the feeling of having had an efficient sleep. For example, assuming you go to bed at 1 o'clock, if you set an alarm to wake up around 5:30 or 40 o'clock and wake up, you can feel refreshed as if you had a short but deep sleep.
하지만, 종래 수면을 깨우는 알람들은 단순히 기상시간을 설정만으로, 수면단계를 고려하지 않고 정해진 시각에 기상알람을 제공한다. 수면주기를 고려하지 않는 기존의 알람 방식은 수면의 질을 저하시킬 수 있다. 또한, 충분한 시간을 수면하였는데도 수면 사이클을 고려하지 않은 기상시간으로 인해 더욱 피곤함을 느끼게 된다.However, conventional sleep alarms simply set a wake-up time and provide a wake-up alarm at a set time without considering the sleep stage. Existing alarm methods that do not consider sleep cycles can reduce sleep quality. In addition, even if you sleep for a sufficient amount of time, you feel more tired due to a wake-up time that does not take into account the sleep cycle.
본 발명의 일 실시예는, 빛과 사운드를 이용하여 뇌파동조를 이룰 수 있도록 하고, 빛과 호흡법을 이용하여 자율신경계를 조절하도록 함으로써, 취침시각 및 기상시각에 맞도록 뇌파 및 자율신경계를 조율하고, 기상시각에 맞춰 스피커를 통하여 세타파, 알파파 및 베타파를 동조하는 모노럴비트 또는 바이노럴비트의 음원을 재생하여 기상시의 뇌파를 유도하며, 기상시각에 블루라이트의 햇빛샤워를 제공하여 각성효과와 함께 세로토닌을 생성하도록 하고, 생성된 세로토닌이 취침시각에 멜라토닌으로 변함으로써 취침 및 숙면을 더욱 유도할 수 있으며, 수동으로 또는 스트레스 지수를 모니터링한 후 임계값을 초과하면 스트레스를 낮추기 위해 교대체감각자극(Bi-Lateral Alternating Stimulation in Tactile)을 사용자 단말 및 웨어러블 기기를 통하여 제공하고, 시상하부 이상활성으로 발생하는 군발두통을 일주기리듬을 정상화시킴으로써 완화시킴과 동시에 비침습성 미주신경자극(NonInvasive Vagus Nerve Stimulation)을 통하여 통증을 완화시킬 수 있도록 하는, AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템을 제공할 수 있다.One embodiment of the present invention enables brain wave synchronization using light and sound, and regulates the autonomic nervous system using light and breathing, thereby adjusting the brain waves and autonomic nervous system to match bedtime and wake-up time. , In accordance with the waking time, a monaural beat or binaural beat sound source that synchronizes theta waves, alpha waves, and beta waves is played through the speaker to induce brain waves when waking up, and a blue light sunlight shower is provided at the waking time to wake up. Along with the effect, it generates serotonin, and the generated serotonin is converted into melatonin at bedtime, which can further induce sleep and sound sleep. Manually or after monitoring the stress index, if it exceeds the threshold, a shift agent is used to lower the stress. Sensory stimulation (Bi-Lateral Alternating Stimulation in Tactile) is provided through user terminals and wearable devices, and cluster headaches caused by abnormal activity of the hypothalamus are alleviated by normalizing the circadian rhythm, while non-invasive vagus nerve stimulation (NonInvasive Vagus) is provided. It is possible to provide a real-time sleep health management service provision system using AI-based brain wave entrainment and autonomic nervous system control to relieve pain through Nerve Stimulation.
또한, 본 발명의 실시예에서는, 스마트 워치와 같은 웨어러블 기기를 통해 사용자가 움직이지 않는 수면중인 상태에서 수집한 사용자별 수면 생체 데이터를 축적하고, 이후, 축적된 수면 생체 데이터를 통해 사용자별 수면 패턴을 분석하고, 사용자별 최적 수면주기 정보를 파악할 수 있는 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템 및 방법을 제공하려는 것을 목적으로 한다.In addition, in an embodiment of the present invention, sleep biometric data for each user collected while the user is sleeping without moving is accumulated through a wearable device such as a smart watch, and then sleep patterns for each user are determined through the accumulated sleep biometric data. The purpose is to provide a system and method for providing real-time sleep health management services using AI-based brain wave entrainment and autonomic nervous system control that can analyze and identify optimal sleep cycle information for each user.
또한, 본 발명의 실시예에서는, 종래 수면을 깨우는 알람이 갖는 문제점(즉, 단순히 기상시간을 설정만으로 수면단계나 수면주기를 고려하지 않고 정해진 시각에 기상알람을 제공함에 따라 사용자의 수면 질을 저하시키고, 또한, 충분한 시간을 수면하였는데도 수면 사이클을 고려하지 않은 기상시간으로 인해 더욱 피곤함을 느끼게 하는 문제 등)을 효과적으로 해소할 수 있는 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템 및 방법을 제공하려는 것을 목적으로 한다.In addition, in an embodiment of the present invention, the user's sleep quality is deteriorated by providing a wake-up alarm at a set time without considering the sleep stage or sleep cycle by simply setting the wake-up time, which has a problem with the conventional alarm that wakes the user from sleep. In addition, we provide a real-time sleep health management service using AI-based brain wave entrainment and autonomic nervous system regulation that can effectively solve problems such as feeling more tired due to waking up time that does not take into account the sleep cycle even after sleeping for a sufficient amount of time. The purpose is to provide systems and methods.
본 발명의 실시예에서, 사용자별 최적 수면주기 정보는 몇 번째 램 수면 시기에 기상했을 때 가장 개운한지 일반화한 것으로서, 실시예에서 제공하는 사용자별 최적 수면주기 정보는 기상시간에 따라 수집된 생체 정보 분석을 통해 사용자 각각에 따라 파악할 수 있다.In an embodiment of the present invention, the optimal sleep cycle information for each user is a generalization of which REM sleep period the user feels most refreshed upon waking up. The optimal sleep cycle information for each user provided in the embodiment is biometric information collected according to the waking time. Through analysis, each user can be identified.
본 발명의 실시예에서는 스마트 워치 등 웨어러블 기기를 이용해, 사용자의 움직임 유무를 분류하고, 움직임 없을 때 심박수 데이터 축적하고, 축적된 데이터 딥러닝을 통해 사용자의 수면단계 예측한다. 이후, 수면단계, 심박수 변화량 고려해 램(REM) 수면단계를 파악하고, 램(REM) 수면 이후 얕은 수면단계에서 스마트 워치를 통해 기상알람을 제공한다. 또한, 실시예에서는 사용자의 움직임과 심박수를 통해 사용자가 기상상태임을 확인할 때까지 알람 반복할 수 있다.In an embodiment of the present invention, wearable devices such as smart watches are used to classify the user's movement, accumulate heart rate data when there is no movement, and predict the user's sleep stage through deep learning of the accumulated data. Afterwards, the REM sleep stage is determined considering the sleep stage and heart rate change, and a wake-up alarm is provided through the smart watch in the light sleep stage after REM sleep. Additionally, in an embodiment, the alarm may be repeated until the user is confirmed to be awake through the user's movements and heart rate.
또한, 본 발명의 실시예에서는 수면에 드는 시간이 평소 패턴(평소 수면 패턴)과 달라지면 전체 수면시간이 달라지게 되므로, 평소 패턴과 다른 시간에 수면을 시작한 경우, 사용자가 기상해야 하는 시간까지 남은 시간을 고려하여 1.5시간의 반복되는 수면주기에 적합한 최적의 기상시간을 재조정하여 사용자를 깨울 수 있도록 한다. In addition, in an embodiment of the present invention, if the time required to sleep is different from the usual pattern (usual sleep pattern), the total sleep time changes, so if sleep starts at a time different from the usual pattern, the time remaining until the time the user must wake up Taking this into account, the optimal wake-up time is readjusted to suit the repeated sleep cycle of 1.5 hours to wake the user.
또한, 본 발명의 실시예에서는 머신 러닝으로 학습한 사용자별 수면 생체 데이터를 통해 잠자리에 누워있는 시간 대비 실제 잠든 시간인 수면효율을 최적화해주는 취침시간과 기상시간 산출하고 이를 취침알람과 기상알람 시간으로 설정할 수 있다. In addition, in an embodiment of the present invention, bedtime and wake-up time that optimize sleep efficiency, which is the actual sleeping time compared to the time lying in bed, are calculated through user-specific sleep biometric data learned through machine learning, and these are converted into bedtime alarm and wake-up alarm times. You can set it.
또한, 본 발명의 실시예에서는 수면단계 판단 및 최적 주기 정보(즉, 최적 수면주기 정보) 파악에 최적화된 맞춤형 트레이닝 데이터 셋(Training Data Set)을 기반으로 딥러닝 뉴럴 네트워크에 기초한 기계학습을 수행하여 최적 기상시간 산출 모델을 구현한다.In addition, in an embodiment of the present invention, machine learning based on a deep learning neural network is performed based on a customized training data set optimized for determining sleep stage and identifying optimal cycle information (i.e., optimal sleep cycle information). Implement an optimal wake-up time calculation model.
다만, 본 실시예가 이루고자 하는 기술적 과제는 상기된 바와 같은 기술적 과제로 한정되지 않으며, 또 다른 기술적 과제들이 존재할 수 있다.However, the technical challenge that this embodiment aims to achieve is not limited to the technical challenges described above, and other technical challenges may exist.
상술한 기술적 과제를 달성하기 위한 기술적 수단으로서, 본 발명의 일 실시예는, 취침시각 및 기상시각을 설정하고, 취침시각에 부교감신경계를 항진시키는 호흡법인 수면유도법을 출력하면서 멜라토닌 비억제 파장의 빛을 조사하며, 기상시각에 알파파 또는 베타파의 뇌파동조를 위하여 알파밴드(Alpha-Band) 또는 베타밴드(Beta-Band)의 플리커(Flicker)를 삽입하여 출력하고, 블루라이트를 출력하여 각성을 유도하는 사용자 단말 및 취침시각 및 기상시각에 뇌파동조 및 자율신경계조절을 위한 빛, 사운드, 진동 및 호흡법에 대한 설정을 저장하는 데이터베이스화부, 사용자 단말로부터 취침시각 및 기상시각을 설정받는 설정부, 사용자 단말에서 설정한 취침시각 및 기상시각에 뇌파동조 및 자율신경계조절을 위한 빛, 사운드, 진동 및 호흡법 중 적어도 하나를 출력하도록 하는 제어부를 포함하는 관리 서비스 제공 서버를 포함한다.As a technical means for achieving the above-mentioned technical problem, one embodiment of the present invention sets the bedtime and wake-up time, outputs a sleep induction method, which is a breathing method that stimulates the parasympathetic nervous system at bedtime, and uses light of a melatonin-inhibiting wavelength. In order to synchronize brain waves with alpha or beta waves at the time of waking up, an alpha-band or beta-band flicker is inserted and output, and blue light is output to promote awakening. A database unit that stores settings for light, sound, vibration and breathing methods for brain wave entrainment and autonomic nervous system control at the user terminal and bedtime and wake-up time, a settings unit that receives settings for bedtime and wake-up time from the user terminal, and the user It includes a management service providing server including a control unit that outputs at least one of light, sound, vibration, and breathing for brain wave entrainment and autonomic nervous system control at the bedtime and wake-up time set in the terminal.
또한, 본 발명의 실시예에서는, 스마트 워치와 같은 웨어러블 기기를 통해 사용자가 움직이지 않는 수면중인 상태에서 수집한 사용자별 수면 생체 데이터를 축적하고, 이후, 축적된 수면 생체 데이터를 통해 사용자별 수면 패턴을 분석하고, 사용자별 최적 수면주기 정보를 파악할 수 있는 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템 및 방법을 제공하려는 것을 목적으로 한다.In addition, in an embodiment of the present invention, sleep biometric data for each user collected while the user is sleeping without moving is accumulated through a wearable device such as a smart watch, and then sleep patterns for each user are determined through the accumulated sleep biometric data. The purpose is to provide a system and method for providing real-time sleep health management services using AI-based brain wave entrainment and autonomic nervous system control that can analyze and identify optimal sleep cycle information for each user.
전술한 본 발명의 과제 해결 수단 중 어느 하나에 의하면, 빛과 사운드를 이용하여 뇌파동조를 이룰 수 있도록 하고, 빛과 호흡법을 이용하여 자율신경계를 조절하도록 함으로써, 취침시각 및 기상시각에 맞도록 뇌파 및 자율신경계를 조율하고, 기상시각에 맞춰 스피커를 통하여 세타파, 알파파 및 베타파를 동조하는 모노럴비트 또는 바이노럴비트의 음원을 재생하여 기상시의 뇌파를 유도하며, 기상시각에 블루라이트의 햇빛샤워를 제공하여 각성효과와 함께 세로토닌을 생성하도록 하고, 생성된 세로토닌이 취침시각에 멜라토닌으로 변함으로써 취침 및 숙면을 더욱 유도할 수 있으며, 수동으로 또는 스트레스 지수를 모니터링한 후 임계값을 초과하면 스트레스를 낮추기 위해 교대체감각자극(Bi-Lateral Alternating Stimulation in Tactile)을 사용자 단말 및 웨어러블 기기를 통하여 제공하고, 시상하부 이상활성으로 발생하는 군발두통을 일주기리듬을 정상화시킴으로써 완화시킴과 동시에 비침습성 미주신경자극(NonInvasive Vagus Nerve Stimulation)을 통하여 통증을 완화시킬 수 있도록 한다.According to one of the means for solving the problems of the present invention described above, brain wave synchronization is achieved using light and sound, and the autonomic nervous system is controlled using light and breathing, so that brain waves are adjusted to match bedtime and waking time. and adjusts the autonomic nervous system, induces brain waves when waking up by playing a sound source of monaural beats or binaural beats that synchronize theta waves, alpha waves, and beta waves through a speaker according to the waking time, and uses blue light at the waking time. By providing a sunlight shower, it generates serotonin along with an awakening effect, and the generated serotonin is converted into melatonin at bedtime to further induce sleep and sound sleep. Manually or after monitoring the stress index, if it exceeds the threshold, To reduce stress, Alternating Somatosensory Stimulation (Bi-Lateral Alternating Stimulation in Tactile) is provided through user terminals and wearable devices, and it relieves cluster headaches caused by abnormal activity of the hypothalamus by normalizing circadian rhythm and is non-invasive. Pain can be alleviated through vagus nerve stimulation (NonInvasive Vagus Nerve Stimulation).
또한, 본 발명의 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템 및 방법은, 수면시간, 취침시간, 기상시간, 낮잠 시간 등을 개인의 수면주기에 따라 사용자의 수면을 맞춤 관리(코칭)함으로써, 수면주기를 이용해 사용자가 짧게 자더라도 개운하게 잘 수 있도록 할 수 있고, 이를 통해, 사용자의 시간 효용성, 수면효율 및 휴식 효율을 높여 삶의 질을 향상시킬 수 있도록 한다. In addition, the system and method for providing a real-time sleep health management service using AI-based brain wave tuning and autonomic nervous system control of the present invention adjusts the user's sleep according to the individual's sleep cycle such as sleep time, bedtime, wake-up time, and nap time. By managing (coaching), the sleep cycle can be used to ensure that the user can sleep refreshed even if he or she sleeps briefly, thereby improving the quality of life by increasing the user's time efficiency, sleep efficiency, and rest efficiency.
또한, 본 발명의 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템 및 방법은, 사용자가 수면 후 충분한 회복이 되지 않은 경우에는 필요한 램(REM) 수면시간을 사용자에게 알려주거나, 사용자 회복을 위한 다양한 서비스를 제공함으로써 사용자의 스트레스 및 피로 회복을 속도를 증진시킬 수 있다.In addition, the system and method for providing real-time sleep health management services using AI-based brain wave entrainment and autonomic nervous system control of the present invention informs the user of the required REM sleep time if the user does not recover sufficiently after sleep, or By providing various services for user recovery, users can speed up their recovery from stress and fatigue.
또한, 본 발명의 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템 및 방법은, 수면단계 판단에 최적화된 맞춤형 트레이닝 데이터 셋(Training Data Set)을 기반으로 딥러닝 뉴럴 네트워크에 기초한 기계학습을 수행하여 수면단계 판단 모델을 구현함으로써, 수면단계 판단 모델로부터 산출되는 수면단계 예측 및 기상시간 산출 결과의 품질을 보다 향상시킬 수 있다.In addition, the system and method for providing real-time sleep health management services using AI-based brain wave entrainment and autonomic nervous system control of the present invention are based on a deep learning neural network based on a customized training data set optimized for sleep stage determination. By implementing a sleep stage judgment model by performing machine learning, the quality of the sleep stage prediction and wake-up time calculation results calculated from the sleep stage judgment model can be further improved.
다만, 본 발명의 효과는 상기한 효과로 한정되는 것은 아니며, 본 발명의 상세한 설명 또는 특허청구범위에 기재된 발명의 구성으로부터 추론 가능한 모든 효과를 포함하는 것으로 이해되어야 한다. 즉, 본 발명에서 얻을 수 있는 효과는 상기된 바와 같은 효과들로 한정되지 않으며, 또 다른 효과들이 존재할 수 있다.However, the effects of the present invention are not limited to the above effects, and should be understood to include all effects that can be inferred from the configuration of the invention described in the detailed description or claims of the present invention. In other words, the effects that can be obtained from the present invention are not limited to the effects described above, and other effects may exist.
도 1은 본 발명의 제 1 실시예에 따른 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템을 설명하기 위한 도면이다.Figure 1 is a diagram for explaining a real-time sleep health management service providing system using AI-based brain wave entrainment and autonomic nervous system control according to the first embodiment of the present invention.
도 2는 도 1의 시스템에 포함된 관리 서비스 제공 서버를 설명하기 위한 블록 구성도이다.FIG. 2 is a block diagram illustrating a management service providing server included in the system of FIG. 1.
도 3 및 도 4는 본 발명의 제 1 실시예에 따른 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스가 구현된 일 실시예를 설명하기 위한 도면이다.Figures 3 and 4 are diagrams for explaining an embodiment in which a real-time sleep health management service using AI-based brain wave entrainment and autonomic nervous system control according to the first embodiment of the present invention is implemented.
도 5는 본 발명의 제 1 실시예에 따른 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 방법을 설명하기 위한 동작 흐름도이다.Figure 5 is an operation flowchart illustrating a method of providing a real-time sleep health management service using AI-based brain wave entrainment and autonomic nervous system control according to the first embodiment of the present invention.
도 6은 일반적인 수면주기를 나타낸 도면이다.Figure 6 is a diagram showing a typical sleep cycle.
도 7은 본 발명의 제 2 실시예에 따른 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템 구성을 나타낸 도면이다.Figure 7 is a diagram showing the configuration of a system for providing real-time sleep health management service using AI-based brain wave entrainment and autonomic nervous system control according to the second embodiment of the present invention.
도 8은 본 발명의 제 2 실시예에 따른 스마트 워치의 데이터 처리 구성을 나타낸 도면이다.Figure 8 is a diagram showing the data processing configuration of a smart watch according to a second embodiment of the present invention.
도 9는 본 발명의 제 2 실시예에 따른 분석 서버의 데이터 처리 구성을 나타낸 도면이다.Figure 9 is a diagram showing the data processing configuration of the analysis server according to the second embodiment of the present invention.
도 10 내지 도 14는 본 발명의 제 2 실시예에 따른 분석 서버에서 이용되는 딥러닝 모델을 설명하기 위한 도면이다.10 to 14 are diagrams for explaining a deep learning model used in an analysis server according to a second embodiment of the present invention.
도 15는 본 발명의 제 2 실시예에 따른 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템의 신호 흐름도이다.Figure 15 is a signal flow diagram of a real-time sleep health management service providing system using AI-based brain wave entrainment and autonomic nervous system control according to the second embodiment of the present invention.
도 16은 본 발명의 제 2 실시예에 따른 기상알람 제공 방법을 나타낸 도면이다.Figure 16 is a diagram showing a method of providing a wake-up alarm according to a second embodiment of the present invention.
도 17은 본 발명의 제 2 실시예에 따른 사용자의 실제 수면시간을 반영한 기상시간 조정 방법을 나타낸 도면이다.Figure 17 is a diagram showing a method of adjusting the wake-up time reflecting the user's actual sleep time according to the second embodiment of the present invention.
아래에서는 첨부한 도면을 참조하여 본 발명이 속하는 기술 분야에서 통상의 지식을 가진 자가 용이하게 실시할 수 있도록 본 발명의 실시예를 상세히 설명한다. 그러나 본 발명은 여러 가지 상이한 형태로 구현될 수 있으며 여기에서 설명하는 실시예에 한정되지 않는다. 그리고 도면에서 본 발명을 명확하게 설명하기 위해서 설명과 관계없는 부분은 생략하였으며, 명세서 전체를 통하여 유사한 부분에 대해서는 유사한 도면 부호를 붙였다.Below, with reference to the attached drawings, embodiments of the present invention will be described in detail so that those skilled in the art can easily implement the present invention. However, the present invention may be implemented in many different forms and is not limited to the embodiments described herein. In order to clearly explain the present invention in the drawings, parts that are not related to the description are omitted, and similar parts are given similar reference numerals throughout the specification.
명세서 전체에서, 어떤 부분이 다른 부분과 "연결"되어 있다고 할 때, 이는 "직접적으로 연결"되어 있는 경우뿐 아니라, 그 중간에 다른 소자를 사이에 두고 "전기적으로 연결"되어 있는 경우도 포함한다. 또한 어떤 부분이 어떤 구성요소를 "포함"한다고 할 때, 이는 특별히 반대되는 기재가 없는 한 다른 구성요소를 제외하는 것이 아니라 다른 구성요소를 더 포함할 수 있는 것을 의미하며, 하나 또는 그 이상의 다른 특징이나 숫자, 단계, 동작, 구성요소, 부분품 또는 이들을 조합한 것들의 존재 또는 부가 가능성을 미리 배제하지 않는 것으로 이해되어야 한다.Throughout the specification, when a part is said to be "connected" to another part, this includes not only the case where it is "directly connected," but also the case where it is "electrically connected" with another element in between. . In addition, when a part is said to "include" a certain component, this does not mean excluding other components unless specifically stated to the contrary, but may further include other components, and one or more other features. It should be understood that it does not exclude in advance the presence or addition of numbers, steps, operations, components, parts, or combinations thereof.
명세서 전체에서 사용되는 정도의 용어 "약", "실질적으로" 등은 언급된 의미에 고유한 제조 및 물질 허용오차가 제시될 때 그 수치에서 또는 그 수치에 근접한 의미로 사용되고, 본 발명의 이해를 돕기 위해 정확하거나 절대적인 수치가 언급된 개시 내용을 비양심적인 침해자가 부당하게 이용하는 것을 방지하기 위해 사용된다. 본 발명의 명세서 전체에서 사용되는 정도의 용어 "~(하는) 단계" 또는 "~의 단계"는 "~ 를 위한 단계"를 의미하지 않는다.The terms "about", "substantially", etc. used throughout the specification are used to mean at or close to that value when manufacturing and material tolerances inherent in the stated meaning are presented, and are used to enhance the understanding of the present invention. Precise or absolute figures are used to assist in preventing unscrupulous infringers from taking unfair advantage of stated disclosures. The term “step of” or “step of” as used throughout the specification of the present invention does not mean “step for.”
본 명세서에 있어서 '부(部)'란, 하드웨어에 의해 실현되는 유닛(unit), 소프트웨어에 의해 실현되는 유닛, 양방을 이용하여 실현되는 유닛을 포함한다. 또한, 1 개의 유닛이 2 개 이상의 하드웨어를 이용하여 실현되어도 되고, 2 개 이상의 유닛이 1 개의 하드웨어에 의해 실현되어도 된다. 한편, '~부'는 소프트웨어 또는 하드웨어에 한정되는 의미는 아니며, '~부'는 어드레싱 할 수 있는 저장 매체에 있도록 구성될 수도 있고 하나 또는 그 이상의 프로세서들을 재생시키도록 구성될 수도 있다. 따라서, 일 예로서 '~부'는 소프트웨어 구성요소들, 객체 지향 소프트웨어 구성요소들, 클래스 구성요소들 및 태스크 구성요소들과 같은 구성요소들과, 프로세스들, 함수들, 속성들, 프로시저들, 서브루틴들, 프로그램 코드의 세그먼트들, 드라이버들, 펌웨어, 마이크로코드, 회로, 데이터, 데이터베이스, 데이터 구조들, 테이블들, 어레이들 및 변수들을 포함한다. 구성요소들과 '~부'들 안에서 제공되는 기능은 더 작은 수의 구성요소들 및 '~부'들로 결합되거나 추가적인 구성요소들과 '~부'들로 더 분리될 수 있다. 뿐만 아니라, 구성요소들 및 '~부'들은 디바이스 또는 보안 멀티미디어카드 내의 하나 또는 그 이상의 CPU들을 재생시키도록 구현될 수도 있다.In this specification, 'part' includes a unit realized by hardware, a unit realized by software, and a unit realized using both. Additionally, one unit may be realized using two or more pieces of hardware, and two or more units may be realized using one piece of hardware. Meanwhile, '~ part' is not limited to software or hardware, and '~ part' may be configured to reside in an addressable storage medium or may be configured to reproduce one or more processors. Therefore, as an example, '~ part' refers to components such as software components, object-oriented software components, class components, and task components, processes, functions, properties, and procedures. , subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays, and variables. The functions provided within the components and 'parts' may be combined into a smaller number of components and 'parts' or may be further separated into additional components and 'parts'. Additionally, components and 'parts' may be implemented to regenerate one or more CPUs within a device or a secure multimedia card.
본 명세서에 있어서 단말, 장치 또는 디바이스가 수행하는 것으로 기술된 동작이나 기능 중 일부는 해당 단말, 장치 또는 디바이스와 연결된 서버에서 대신 수행될 수도 있다. 이와 마찬가지로, 서버가 수행하는 것으로 기술된 동작이나 기능 중 일부도 해당 서버와 연결된 단말, 장치 또는 디바이스에서 수행될 수도 있다.In this specification, some of the operations or functions described as being performed by a terminal, apparatus, or device may instead be performed on a server connected to the terminal, apparatus, or device. Likewise, some of the operations or functions described as being performed by the server may also be performed in a terminal, apparatus, or device connected to the server.
본 명세서에서 있어서, 단말과 매핑(Mapping) 또는 매칭(Matching)으로 기술된 동작이나 기능 중 일부는, 단말의 식별 정보(Identifying Data)인 단말기의 고유번호나 개인의 식별정보를 매핑 또는 매칭한다는 의미로 해석될 수 있다.In this specification, some of the operations or functions described as mapping or matching with the terminal mean mapping or matching the terminal's unique number or personal identification information, which is identifying data of the terminal. It can be interpreted as
이하 첨부된 도면을 참고하여 본 발명을 상세히 설명하기로 한다.Hereinafter, the present invention will be described in detail with reference to the attached drawings.
도 1은 본 발명의 제 1 실시예에 따른 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템을 설명하기 위한 도면이다. 도 1을 참조하면, AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템(1)은, 적어도 하나의 사용자 단말(100), 관리 서비스 제공 서버(300), 적어도 하나의 웨어러블 기기(400)를 포함할 수 있다. 다만, 이러한 도 1의 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템(1)은, 본 발명의 일 실시예에 불과하므로, 도 1을 통하여 본 발명이 한정 해석되는 것은 아니다.Figure 1 is a diagram for explaining a real-time sleep health management service providing system using AI-based brain wave entrainment and autonomic nervous system control according to the first embodiment of the present invention. Referring to FIG. 1, a real-time sleep health management service providing system 1 using AI-based brain wave tuning and autonomic nervous system control includes at least one user terminal 100, a management service providing server 300, and at least one wearable device. It may include (400). However, the real-time sleep health management service providing system (1) using AI-based brain wave tuning and autonomic nervous system control shown in FIG. 1 is only an embodiment of the present invention, and the present invention is not limited to FIG. 1. .
이때, 도 1의 각 구성요소들은 일반적으로 네트워크(Network, 200)를 통해 연결된다. 예를 들어, 도 1에 도시된 바와 같이, 적어도 하나의 사용자 단말(100)은 네트워크(200)를 통하여 관리 서비스 제공 서버(300)와 연결될 수 있다. 그리고, 관리 서비스 제공 서버(300)는, 네트워크(200)를 통하여 적어도 하나의 사용자 단말(100), 적어도 하나의 웨어러블 기기(400)와 연결될 수 있다. 또한, 적어도 하나의 웨어러블 기기(400)는, 네트워크(200)를 통하여 관리 서비스 제공 서버(300)와 연결될 수 있다. At this time, each component of FIG. 1 is generally connected through a network (Network, 200). For example, as shown in FIG. 1, at least one user terminal 100 may be connected to the management service providing server 300 through the network 200. In addition, the management service providing server 300 may be connected to at least one user terminal 100 and at least one wearable device 400 through the network 200. Additionally, at least one wearable device 400 may be connected to the management service providing server 300 through the network 200.
여기서, 네트워크는, 복수의 단말 및 서버들과 같은 각각의 노드 상호 간에 정보 교환이 가능한 연결 구조를 의미하는 것으로, 이러한 네트워크의 일 예에는 근거리 통신망(LAN: Local Area Network), 광역 통신망(WAN: Wide Area Network), 인터넷(WWW: World Wide Web), 유무선 데이터 통신망, 전화망, 유무선 텔레비전 통신망 등을 포함한다. 무선 데이터 통신망의 일례에는 3G, 4G, 5G, 3GPP(3rd Generation Partnership Project), 5GPP(5th Generation Partnership Project), 5G NR(New Radio), 6G(6th Generation of Cellular Networks), LTE(Long Term Evolution), WIMAX(World Interoperability for Microwave Access), 와이파이(Wi-Fi), 인터넷(Internet), LAN(Local Area Network), Wireless LAN(Wireless Local Area Network), WAN(Wide Area Network), PAN(Personal Area Network), RF(Radio Frequency), 블루투스(Bluetooth) 네트워크, NFC(Near-Field Communication) 네트워크, 위성 방송 네트워크, 아날로그 방송 네트워크, DMB(Digital Multimedia Broadcasting) 네트워크 등이 포함되나 이에 한정되지는 않는다.Here, the network refers to a connection structure that allows information exchange between each node, such as a plurality of terminals and servers. Examples of such networks include a local area network (LAN) and a wide area network (WAN). Wide Area Network, Internet (WWW: World Wide Web), wired and wireless data communication network, telephone network, wired and wireless television communication network, etc. Examples of wireless data communication networks include 3G, 4G, 5G, 3rd Generation Partnership Project (3GPP), 5th Generation Partnership Project (5GPP), 5G New Radio (NR), 6th Generation of Cellular Networks (6G), and Long Term Evolution (LTE). , WIMAX (World Interoperability for Microwave Access), Wi-Fi, Internet, LAN (Local Area Network), Wireless LAN (Wireless Local Area Network), WAN (Wide Area Network), PAN (Personal Area Network) ), RF (Radio Frequency), Bluetooth (Bluetooth) network, NFC (Near-Field Communication) network, satellite broadcasting network, analog broadcasting network, DMB (Digital Multimedia Broadcasting) network, etc., but are not limited to these.
하기에서, 적어도 하나의 라는 용어는 단수 및 복수를 포함하는 용어로 정의되고, 적어도 하나의 라는 용어가 존재하지 않더라도 각 구성요소가 단수 또는 복수로 존재할 수 있고, 단수 또는 복수를 의미할 수 있음은 자명하다 할 것이다. 또한, 각 구성요소가 단수 또는 복수로 구비되는 것은, 실시예에 따라 변경가능하다 할 것이다.In the following, the term at least one is defined as a term including singular and plural, and even if the term at least one does not exist, each component may exist in singular or plural, and may mean singular or plural. This should be self-explanatory. In addition, whether each component is provided in singular or plural form may be changed depending on the embodiment.
적어도 하나의 사용자 단말(100)은, AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 관련 웹 페이지, 앱 페이지, 프로그램 또는 애플리케이션을 이용하여 취침시각 및 기상시각을 설정하고, 이에 따른 빛, 사운드, 진동 및 호흡법을 출력하는 사용자의 단말일 수 있다.At least one user terminal 100 sets bedtime and wake-up time using a web page, app page, program, or application related to a real-time sleep health management service using AI-based brain wave entrainment and autonomic nervous system control, and emits light accordingly. , It may be a user terminal that outputs sound, vibration, and breathing techniques.
여기서, 적어도 하나의 사용자 단말(100)은, 네트워크를 통하여 원격지의 서버나 단말에 접속할 수 있는 컴퓨터로 구현될 수 있다. 여기서, 컴퓨터는 예를 들어, 네비게이션, 웹 브라우저(WEB Browser)가 탑재된 노트북, 데스크톱(Desktop), 랩톱(Laptop) 등을 포함할 수 있다. 이때, 적어도 하나의 사용자 단말(100)은, 네트워크를 통해 원격지의 서버나 단말에 접속할 수 있는 단말로 구현될 수 있다. 적어도 하나의 사용자 단말(100)은, 예를 들어, 휴대성과 이동성이 보장되는 무선 통신 장치로서, 네비게이션, PCS(Personal Communication System), GSM(Global System for Mobile communications), PDC(Personal Digital Cellular), PHS(Personal Handyphone System), PDA(Personal Digital Assistant), IMT(International Mobile Telecommunication)-2000, CDMA(Code Division Multiple Access)-2000, W-CDMA(W-Code Division Multiple Access), Wibro(Wireless Broadband Internet) 단말, 스마트폰(Smartphone), 스마트 패드(Smartpad), 타블렛 PC(Tablet PC) 등과 같은 모든 종류의 핸드헬드(Handheld) 기반의 무선 통신 장치를 포함할 수 있다.Here, at least one user terminal 100 may be implemented as a computer capable of accessing a remote server or terminal through a network. Here, the computer may include, for example, a laptop equipped with a navigation system and a web browser, a desktop, a laptop, etc. At this time, at least one user terminal 100 may be implemented as a terminal capable of accessing a remote server or terminal through a network. At least one user terminal 100 is, for example, a wireless communication device that guarantees portability and mobility, and includes navigation, personal communication system (PCS), global system for mobile communications (GSM), personal digital cellular (PDC), PHS (Personal Handyphone System), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), Wibro (Wireless Broadband Internet) ) It may include all types of handheld-based wireless communication devices such as terminals, smartphones, smartpads, and tablet PCs.
관리 서비스 제공 서버(300)는, AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 웹 페이지, 앱 페이지, 프로그램 또는 애플리케이션을 제공하는 서버일 수 있다. 그리고, 관리 서비스 제공 서버(300)는, 빛, 사운드, 진동 및 호흡법에 따른 설정을 데이터베이스화하는 서버일 수 있다. 또한, 관리 서비스 제공 서버(300)는, 사용자 단말(100)에서 취침시각 및 기상시각에 뇌파동조 및 자율신경계조절을 위한 빛, 사운드, 진동 및 호흡법 중 적어도 하나를 출력하도록 하는 서버일 수 있다. 그리고, 관리 서비스 제공 서버(300)는, 교대체감각자극(Bi-Lateral Alternating Stimulation in Tactile)을 사용자 단말(100) 및 웨어러블 기기(400)를 통하여 제공할 수 있도록 하고, 비침습성 미주신경자극(NonInvasive Vagus Nerve Stimulation)을 통하여 통증을 완화시킬 수 있도록 미주신경자극기를 제어하거나 사용자 단말(100)로 고주파를 출력하도록 하는 서버일 수 있다.The management service providing server 300 may be a server that provides a real-time sleep health management service web page, app page, program, or application using AI-based brain wave entrainment and autonomic nervous system control. And, the management service providing server 300 may be a server that databases settings according to light, sound, vibration, and breathing methods. Additionally, the management service providing server 300 may be a server that allows the user terminal 100 to output at least one of light, sound, vibration, and breathing methods for brain wave entrainment and autonomic nervous system control at bedtime and waking time. In addition, the management service providing server 300 allows bi-lateral somatosensory stimulation (Bi-Lateral Alternating Stimulation in Tactile) to be provided through the user terminal 100 and the wearable device 400, and non-invasive vagus nerve stimulation ( It may be a server that controls the vagus nerve stimulator to relieve pain through NonInvasive Vagus Nerve Stimulation or outputs high frequencies to the user terminal 100.
여기서, 관리 서비스 제공 서버(300)는, 네트워크를 통하여 원격지의 서버나 단말에 접속할 수 있는 컴퓨터로 구현될 수 있다. 여기서, 컴퓨터는 예를 들어, 네비게이션, 웹 브라우저(WEB Browser)가 탑재된 노트북, 데스크톱(Desktop), 랩톱(Laptop) 등을 포함할 수 있다.Here, the management service providing server 300 may be implemented as a computer that can connect to a remote server or terminal through a network. Here, the computer may include, for example, a laptop equipped with a navigation system and a web browser, a desktop, a laptop, etc.
적어도 하나의 웨어러블 기기(400)는, AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 관련 웹 페이지, 앱 페이지, 프로그램 또는 애플리케이션을 이용하는 수면단계를 파악할 수 있도록 하는 감지 데이터를 업로드하고, 기 설정된 주파수의 진동을 기 설정된 강도, 시간에 따라 출력하는 장치일 수 있다.At least one wearable device 400 uploads sensing data that allows the user to determine the sleep stage using a web page, app page, program, or application related to a real-time sleep health management service using AI-based brain wave entrainment and autonomic nervous system control, It may be a device that outputs vibration of a preset frequency according to a preset intensity and time.
여기서, 적어도 하나의 웨어러블 기기(400)는, 네트워크를 통하여 원격지의 서버나 단말에 접속할 수 있는 컴퓨터로 구현될 수 있다. 여기서, 컴퓨터는 예를 들어, 네비게이션, 웹 브라우저(WEB Browser)가 탑재된 노트북, 데스크톱(Desktop), 랩톱(Laptop) 등을 포함할 수 있다. 이때, 적어도 하나의 웨어러블 기기(400)는, 네트워크를 통해 원격지의 서버나 단말에 접속할 수 있는 단말로 구현될 수 있다. 적어도 하나의 웨어러블 기기(400)는, 예를 들어, 휴대성과 이동성이 보장되는 무선 통신 장치로서, 네비게이션, PCS(Personal Communication System), GSM(Global System for Mobile communications), PDC(Personal Digital Cellular), PHS(Personal Handyphone System), PDA(Personal Digital Assistant), IMT(International Mobile Telecommunication)-2000, CDMA(Code Division Multiple Access)-2000, W-CDMA(W-Code Division Multiple Access), Wibro(Wireless Broadband Internet) 단말, 스마트폰(Smartphone), 스마트 패드(Smartpad), 타블렛 PC(Tablet PC) 등과 같은 모든 종류의 핸드헬드(Handheld) 기반의 무선 통신 장치를 포함할 수 있다.Here, at least one wearable device 400 may be implemented as a computer capable of accessing a remote server or terminal through a network. Here, the computer may include, for example, a laptop equipped with a navigation system and a web browser, a desktop, a laptop, etc. At this time, at least one wearable device 400 may be implemented as a terminal capable of accessing a remote server or terminal through a network. At least one wearable device 400 is, for example, a wireless communication device that ensures portability and mobility, and includes navigation, personal communication system (PCS), global system for mobile communications (GSM), personal digital cellular (PDC), PHS (Personal Handyphone System), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), Wibro (Wireless Broadband Internet) ) It may include all types of handheld-based wireless communication devices such as terminals, smartphones, smartpads, and tablet PCs.
도 2는 도 1의 시스템에 포함된 관리 서비스 제공 서버를 설명하기 위한 블록 구성도이고, 도 3 및 도 4는 본 발명의 일 실시예에 따른 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스가 구현된 일 실시예를 설명하기 위한 도면이다.Figure 2 is a block diagram for explaining the management service providing server included in the system of Figure 1, and Figures 3 and 4 are real-time sleep health using AI-based brain wave entrainment and autonomic nervous system control according to an embodiment of the present invention. This is a diagram to explain an embodiment in which a management service is implemented.
도 2를 참조하면, 관리 서비스 제공 서버(300)는, 데이터베이스화부(310), 설정부(320), 제어부(330), 기상뇌파유도부(340), 수면단계별코칭부(350), 인공지능부(360), 불면과다완화부(370), 햇빛샤워부(380), 일주기리듬맞춤부(390), 미주신경자극부(391) 및 교대체감각자극부(393)를 포함할 수 있다.Referring to FIG. 2, the management service providing server 300 includes a database unit 310, a setting unit 320, a control unit 330, a wake-up brain wave induction unit 340, a sleep stage coaching unit 350, and an artificial intelligence unit. It may include (360), an insomnia excessive relief unit (370), a sunlight shower unit (380), a circadian rhythm adjustment unit (390), a vagus nerve stimulation unit (391), and an alternating somatosensory stimulation unit (393).
본 발명의 일 실시예에 따른 관리 서비스 제공 서버(300)나 연동되어 동작하는 다른 서버(미도시)가 적어도 하나의 사용자 단말(100) 및 적어도 하나의 웨어러블 기기(400)로 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 애플리케이션, 프로그램, 앱 페이지, 웹 페이지 등을 전송하는 경우, 적어도 하나의 사용자 단말(100), 적어도 하나의 웨어러블 기기(400) 및 적어도 하나의 정보제공 서버(500)는, AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 애플리케이션, 프로그램, 앱 페이지, 웹 페이지 등을 설치하거나 열 수 있다. 또한, 웹 브라우저에서 실행되는 스크립트를 이용하여 서비스 프로그램이 적어도 하나의 사용자 단말(100), 적어도 하나의 웨어러블 기기(400) 및 적어도 하나의 정보제공 서버(500)에서 구동될 수도 있다. 여기서, 웹 브라우저는 웹(WWW: World Wide Web) 서비스를 이용할 수 있게 하는 프로그램으로 HTML(Hyper Text Mark-up Language)로 서술된 하이퍼 텍스트를 받아서 보여주는 프로그램을 의미하며, 예를 들어 크롬(Chrome), 에지(Microsoft Edge), 사파리(Safari), 파이어폭스(FireFox), 웨일(Whale), UC 브라우저 등을 포함한다. 또한, 애플리케이션은 단말 상의 응용 프로그램(Application)을 의미하며, 예를 들어, 모바일 단말(스마트폰)에서 실행되는 앱(App)을 포함한다.The management service providing server 300 according to an embodiment of the present invention or another server (not shown) operating in conjunction with at least one user terminal 100 and at least one wearable device 400 performs AI-based brain wave entrainment and When transmitting a real-time sleep health management service application, program, app page, web page, etc. using autonomic nervous system regulation, at least one user terminal 100, at least one wearable device 400, and at least one information providing server ( 500) can install or open real-time sleep health management service applications, programs, app pages, web pages, etc. using AI-based brain wave tuning and autonomic nervous system control. Additionally, a service program may be run on at least one user terminal 100, at least one wearable device 400, and at least one information provision server 500 using a script executed in a web browser. Here, a web browser is a program that allows the use of web (WWW: World Wide Web) services and refers to a program that receives and displays hypertext written in HTML (Hyper Text Mark-up Language), for example, Chrome. , Microsoft Edge, Safari, FireFox, Whale, UC Browser, etc. Additionally, an application refers to an application on a terminal and includes, for example, an app running on a mobile terminal (smartphone).
도 2를 참조하면, 데이터베이스화부(310)는, 취침시각 및 기상시각에 뇌파동조 및 자율신경계조절을 위한 빛, 사운드, 진동 및 호흡법에 대한 설정을 저장할 수 있다. 우선, 뇌파란, 뇌가 활성화되면 신경세포의 미세전류에서 보강간섭으로 인해 측정되는 생체 정보이다. 한 위치에서 연속적인 데이터를 측정하며 파동의 특징인 진폭과 주파수, 파형을 가지고 있다. 뇌파는 신경세포의 활성화 상태에서 다른 특징을 보이기 때문에 뇌파를 이용하여 뇌의 활동성을 측정할 수 있다. 그래서 수면다원검사에서는 뇌파를 이용하여 수면의 깊이를 평가한다. 사람은 수면상태에서 무의식으로 신체활동이 감소하기 때문에 수면상태에서 측정한 수면뇌파는 각성(비수면)상태에서 측정한 뇌파와 다른 특징을 보인다. 수면 뇌파의 특징을 이용하여 수면의 깊이인 수면단계를 구분하기 위해 수면기록 매뉴얼(Rechtschaffen & Kales Sleep Scoring Manual)에서 정의된 수면단계를 이용한다. 이 수면단계 기준은 수면다원검사를 포함하여 다양한 연구 분야에서 사용되고 있으며 이하 표 1과 같다.Referring to FIG. 2, the database unit 310 can store settings for light, sound, vibration, and breathing methods for brain wave entrainment and autonomic nervous system control at bedtime and waking time. First of all, brain waves are biometric information measured due to constructive interference in the microcurrents of nerve cells when the brain is activated. It measures continuous data at one location and has amplitude, frequency, and waveform, which are characteristics of a wave. Because brain waves show different characteristics depending on the state of nerve cell activation, brain activity can be measured using brain waves. Therefore, polysomnography evaluates the depth of sleep using brain waves. Because people's physical activity decreases unconsciously in a sleep state, sleep EEGs measured in a sleep state show different characteristics from EEGs measured in an awake (non-sleep) state. To classify the sleep stage, which is the depth of sleep, using the characteristics of sleep EEG, the sleep stage defined in the sleep recording manual (Rechtschaffen & Kales Sleep Scoring Manual) is used. This sleep stage standard is used in various research fields, including polysomnography, and is shown in Table 1 below.
| 단계step | 주파수 특징frequency characteristics | 기타 특징Other Features |
| 각성(wake)wake up | 알파파(8-13Hz) 우세Alpha waves (8-13Hz) dominate | 움직임과 심박으로 인한 노이즈 발생Noise caused by movement and heart rate |
| 1One | 혼합파(2-7Hz) 우세, 알파파 50% 이하Mixed waves (2-7Hz) dominate, alpha waves less than 50% |
K-복합체와 수면방추파 없음No K-complexes and |
| 22 | 서파(0-2Hz)가 20% 이하Slow waves (0-2Hz) less than 20% | K-복합체와 수면방추파 측정됨K-complex and sleep spindle waves measured |
| 33 | 서파가 20%에서 50% 측정됨Slow waves measured in 20% to 50% of cases | -- |
| 44 | 서파가 50% 이상 측정됨Slow waves measured in more than 50% of cases | -- |
| REMR.E.M. | 혼합파 우세Mixed faction dominates | 안구 움직임으로 인한 노이즈 발생Noise caused by eye movements |
각성단계(Wake)는 비수면상태로 알파파(8-12Hz)가 활성화되며 움직임과 심박에 의한 노이즈가 발생한다. 1 단계 수면은 2-7Hz의 혼합된 뇌파가 측정되며 K복합체와 수면방추파가 검출되지 않는다. K복합체는 일시적으로 나타나는 높은 진폭의 양성 방향 파형을 의미하고 수면방추파는 K 복합체 뒤에 짧게 측정되는 12-14 Hz의 복합파이다. 2 단계 수면에서는 K 복합체와 수면 방추파가 측정되고 2Hz 이하의 느린 파형이 20% 이하로 구성된다. 3 단계 수면은 2Hz 이하의 느린 뇌파가 20%에서 50%로 측정되면 3단계로 구분된다. 4 단계 수면은 2Hz 이하의 느린 뇌파가 50% 이상 포함된다. REM 수면은 1 단계 수면과 비슷하지만 안구 움직임으로 인한 노이즈가 같이 측정된다. 이때 뇌파의 종류 및 주파수는 이하 표 2와 같다.The wake stage is a non-sleep state, where alpha waves (8-12 Hz) are activated and noise is generated due to movement and heart rate. In stage 1 sleep, mixed brain waves of 2-7Hz are measured, and K complexes and sleep spindle waves are not detected. K-complexes are transient, high-amplitude, positive-directed waves, and sleep spindles are short, 12-14 Hz complex waves that follow the K complexes. In stage 2 sleep, K complexes and sleep spindle waves are measured, and slow waves of 2 Hz or less make up less than 20%. Stage 3 sleep is classified into three stages when 20% to 50% of the brain waves are slower than 2Hz. Stage 4 sleep includes more than 50% of slow brain waves below 2Hz. REM sleep is similar to stage 1 sleep, but noise caused by eye movements is also measured. At this time, the types and frequencies of brain waves are shown in Table 2 below.
| 뇌파의 종류Types of brain waves | 주파수 대역frequency band | 뇌의 상태condition of the brain |
| 델타(Delta)Delta | 0.5~4 Hz0.5~4 Hz |
숙면 상태deep |
| 세타(Theta)Theta | 4~7 Hz4~7 Hz |
졸리는 상태, 산만함, 백일몽 상태Sleepy state, distracted state, daydreaming |
| 알파(Alpha)Alpha | 8~12Hz8~12Hz | 편안한 상태에서 외부 집중력이 느슨한 상태A state of relaxed external concentration |
| SMR(Sensory Motor Rhythm)Sensory Motor Rhythm (SMR) | 12~15Hz12~15Hz |
움직이지 않는 상태에서 집중력을 유지긴장과 이완의 중간상태Maintain concentration without moving A state between tension and |
| 베타(Beta)Beta | 15~30Hz15~30Hz | 사고를 하며 활동적인 상태에서 집중력을 유지Stay focused while thinking and being active |
| 감마(Gamma)Gamma | 31~50Hz31~50Hz | 피질과 피질 하 영역 간 정보교환의식적 각성상태와 REM 수면시 꿈에서 나타남베타파와 중복되어 나타나기도 함Information exchange between cortical and subcortical areas Appears in dreams during conscious awakening and REM sleep May overlap with beta waves |
<뇌파동조>뇌파동조이론(The Frequence-Following Effect)은 뇌전도 동기화에 따른 신경동조라고도 하는데, 깜박이는 조명, 음성, 음악 및 음악과 같은 주기적인 외부 자극의 리듬에 의해 뇌파(뇌의 큰 전기 진동)가 원하는 주파수로 맞춰진다는 이론이다. 이는, V. A. Korshunov, G. R. Khazankin and D. S. Ivanishkin, "Development of an Application for Audio-Visual-Tactile Brainwave Entrainment in Patients with Affective and Psychosomatic Disorders," 2021 IEEE 22nd International Conference of Young Professionals in Electron Devices and Materials(EDM), Souzga, the Altai Republic, Russia, 2021, pp. 551-554, doi: 10.1109/EDM52169.2021.9507617나, H. Norhazman, N. Zaini, M. N. Taib, R. Jailani and M. F. A. Latip, "Alpha and Beta Sub-waves Patterns when Evoked by External Stressor and Entrained by Binaural Beats Tone," 2019 IEEE 7th Conference on Systems, Process and Control(ICSPC), Melaka, Malaysia, 2019, pp. 112-117, doi: 10.1109/ICSPC47137.2019.9068008 등에 게재되어 있다. 본 발명의 일 실시예에서는 뇌파동조이론에 기반하여 빛(알파밴드 및 베타밴드)과 사운드(모노럴비트 및 바이노럴비트)를 이용하여 수면에 접어들게 하며, 블루라이트를 이용하여 각성을 유도할 수 있도록 한다.<Brain wave entrainment> Brain wave tuning theory (The Frequence-Following Effect) is also called neural entrainment based on electroencephalogram synchronization, and is caused by brain waves (large electrical oscillations in the brain) caused by the rhythm of periodic external stimuli such as flashing lights, voices, music, and music. ) is adjusted to the desired frequency. This is V. A. Korshunov, G. R. Khazankin and D. S. Ivanishkin, "Development of an Application for Audio-Visual-Tactile Brainwave Entrainment in Patients with Affective and Psychosomatic Disorders," 2021 IEEE 22nd International Conference of Young Professionals in Electron Devices and Materials (EDM), Souzga, the Altai Republic, Russia, 2021, pp. 551-554, doi: 10.1109/EDM52169.2021.9507617I, H. Norhazman, N. Zaini, M. N. Taib, R. Jailani and M. F. A. Latip, "Alpha and Beta Sub-waves Patterns when Evoked by External Stressor and Entrained by Binaural Beats Tone ," 2019 IEEE 7th Conference on Systems, Process and Control (ICSPC), Melaka, Malaysia, 2019, pp. 112-117, doi: 10.1109/ICSPC47137.2019.9068008, etc. In one embodiment of the present invention, based on brain wave entrainment theory, light (alpha band and beta band) and sound (monaural beat and binaural beat) are used to enter sleep, and blue light is used to induce awakening. make it possible
<자율신경계조절><Autonomous nervous system control>
교감 신경의 경우 척수의 중간 부분에서 나와 여러 내장기관에 분포하며 위급한 상황에 빠졌을 경우 빠르게 대처할 수 있도록 도와주는 역할을 한다. 교감신경 흥분 시 동공은 확장되고 땀의 분비가 촉진되며 심장박동수가 증가하고 혈관은 수축하며, 기관지가 확장되며 위장관 운동은 저하된다. 반대로, 부교감 신경은, 중뇌와 연수 및 척수의 꼬리부분에서 나와 각 내장기관에 분포하며 위급한 상황에 대비하여 에너지를 저장해두는 역할을 한다. 부교감신경 흥분 시 동공은 수축하고 땀 분비는 감소하며 심박동수는 감소하고 일부 혈관이 확장될 수 있다. 또한 기관지는 수축하며 위장관 운동이 촉진됩니다. 이에 교감 신경은 각성을 유도하고 반대로 부교감 신경은 수면을 유도하게 되므로, 수면을 유도할 때에는 멜라토닌을 유도하는 적색 및 황색의 빛을 제공하고, 호흡법을 도 4e 및 도 4f와 같은 깊은 호흡법으로 유도하여 부교감신경계를 항진시킬 수 있도록 한다. 이때, 적색 및 황색의 빛이 멜라토닌을 유도한다는 이론적 근거는 논문(Blume C, Garbazza C, Spitschan M. Effects of light on human circadian rhythms, sleep and mood. Somnologie (Berl). 2019 Sep;23(3):147-156. doi: 10.1007/s11818-019-00215-x. Epub 2019 Aug 20. PMID: 31534436; PMCID: PMC6751071.) 및 도 4d에 근거한다. 각성시에는 상술한 방법과 반대의 상황을 유도할 수 있다.In the case of the sympathetic nerve, it comes out of the middle part of the spinal cord and is distributed to various internal organs, and plays a role in helping people respond quickly in case of an emergency. When sympathetic nerves are excited, pupils dilate, sweat secretion is promoted, heart rate increases, blood vessels constrict, bronchi dilate, and gastrointestinal motility decreases. On the other hand, the parasympathetic nerves come out of the midbrain, medulla, and tail of the spinal cord, are distributed to each internal organ, and play the role of storing energy in preparation for emergency situations. During parasympathetic stimulation, pupils constrict, sweat secretion decreases, heart rate decreases, and some blood vessels may dilate. Additionally, the bronchi constrict and gastrointestinal motility is promoted. Accordingly, the sympathetic nerve induces awakening and, conversely, the parasympathetic nerve induces sleep. Therefore, when inducing sleep, red and yellow light that induces melatonin is provided, and breathing is induced into a deep breathing method as shown in Figures 4e and 4f. Helps stimulate the parasympathetic nervous system. At this time, the theoretical basis that red and yellow light induces melatonin is found in the paper (Blume C, Garbazza C, Spitschan M. Effects of light on human circadian rhythms, sleep and mood. Somnologie (Berl). 2019 Sep;23(3) :147-156. doi: 10.1007/s11818-019-00215-x. Epub 2019 Aug 20. PMID: 31534436; PMCID: PMC6751071.) and based on Fig. 4d. Upon awakening, a situation opposite to the method described above can be induced.
<플리커 삽입><insert flicker>
기상시각 이전부터 사용자 단말(100)의 디스플레이를 점점 밝게 하면서 알람시간에 가까워질수록 최대 밝기로 디스플레이와 플래시 라이트를 점등할 수 있다. 이때 애플리케이션 소프트웨어와 연동하여 LED 장치, 디스플레이, 플래시 라이트로부터 알파밴드 또는 베타밴드의 플리커, 즉, 영상 콘텐츠에 깜박이는 효과를 줄 수 있는 이미지 프레임을 삽입할 수 있다. 플리커는, 디스플레이 주사율 조절, 디스플레이, 플래시 라이트 및 연동되는 LED를 점멸시킴으로써 발생시킬 수도 있다. 이에 플리커를 이용하여 알파파 또는 베타파를 동조시킬 수 있다. 이 이론은 상술한 논문을 기반으로 한다.The display of the user terminal 100 can be gradually brightened before the wake-up time, and the display and flash light can be turned on at maximum brightness as the alarm time approaches. At this time, in conjunction with the application software, it is possible to insert alpha-band or beta-band flicker from an LED device, display, or flash light, that is, an image frame that can give a flickering effect to video content. Flicker can also be caused by adjusting the display refresh rate and blinking the display, flash light, and associated LEDs. Accordingly, flicker can be used to synchronize alpha waves or beta waves. This theory is based on the above-mentioned paper.
<호흡법><Breathing method>
도 4e와 같이 사용자 단말(100)의 디스플레이를 통해 도형의 크기 변화와 숫자를 이용한 호흡법을 코칭하여 뇌파변화를 유도시킬 수 있다. 이 근거는 논문(강승완.(2017).의식적 호흡이 자율신경과 뇌파에 영향을 미치는 기전에 관하여.Perspectives in Nursing Science,14(2),64-69.)에 기반한다. 이때, 심장박동과 같은 자율신경계가 의식적 호흡변화를 통해 조절이 가능한데, 예를 들어 화가 많이 나서 교감신경이 극도로 흥분한 사람에게 심호흡을 하라는 이유도 여기에 있다. 의식적으로 호흡으로 심장박동수를 변화시키라는 것이다. 이러한 심장의 리듬패턴의 변화가 뇌파의 변화를 이끌어낼 수 있다. 이때 도 4f와 같이 사용자별 호흡법 조절 사용자 인터페이스도 제공할 수 있는데, 호흡단계별 비율은 유지하되 호흡법의 숙련 정도에 따라 속도, 횟수, 세트를 조절하여 사용자별로 적합한 호흡법을 선택할 수 있도록 할 수 있다. 이 호흡법은 비교적 안전하지만 처음 연습시에는 약간의 어지러움을 느낄 수 있다. 정상적 호흡은 산소를 들이마시는 것과 이산화탄소를 내쉬는 것 사이의 균형인데, 숨을 들이마신 것보다 더 많이 내쉬면서 이 균형을 깨뜨릴 때, 체내 이산화탄소의 감소를 초래한다. 이로 인해 뇌에 혈액을 공급하는 혈관이 좁아지면서, 혈액 공급의 감소가 어지럼증 같은 증상으로 이어질 수 있다. 따라서 천천히 연습하는 기간을 조절할 수 있도록 하고, 호흡에 익숙해지면 횟수를 늘리도록 할 수 있다. 또, 눈감고 호흡하기를 유도할 수 있는데, 어두운 곳에서 눈을 감고 빛을 느끼며 호흡할 수 있도록 디스플레이의 밝기와 색상이 변환되고 깜빡일 수 있도록 한다. 예를 들어, 도 4e와 같이 어두운 밝기의 검은색에서 숨을 들이쉴수록 점점 밝아지는 붉은색을 점등할 수 있다. 또, 도 4g와 같이 [숨 들이쉬기(디스플레이가 최소 밝기의 검은색에서 점점 최대 밝기의 붉은색으로 전환)→숨 참기(디스플레이가 깜빡임)→숨 내쉬기(화면 밝기와 색상이 점점 어두워져 최소 밝기로 전환)→세트 사이 휴식(디스플레이 밝기와 색상 유지)]로 세팅할 수도 있다.As shown in FIG. 4E, brain wave changes can be induced by coaching breathing methods using changes in the size of shapes and numbers through the display of the user terminal 100. This evidence is based on the paper (Seungwan Kang (2017). On the mechanism by which conscious breathing affects the autonomic nervous system and brain waves. Perspectives in Nursing Science, 14(2), 64-69.). At this time, the autonomic nervous system, such as heart rate, can be controlled through conscious breathing changes. For example, this is why a person who is very angry and whose sympathetic nervous system is extremely excited is asked to take deep breaths. The idea is to consciously change your heart rate through breathing. These changes in the heart's rhythm pattern can lead to changes in brain waves. At this time, a user interface for controlling breathing methods for each user can be provided as shown in Figure 4f. The ratio of each breathing step can be maintained, but the speed, number of times, and sets can be adjusted according to the level of proficiency in the breathing method to select a breathing method suitable for each user. This breathing technique is relatively safe, but you may feel a little dizzy when you first practice it. Normal breathing is a balance between inhaling oxygen and exhaling carbon dioxide, but when this balance is disrupted by exhaling more than inhaling, it results in a decrease in carbon dioxide in the body. This causes the blood vessels that supply blood to the brain to narrow, and the decrease in blood supply can lead to symptoms such as dizziness. Therefore, you can slowly adjust the period of practice and increase the number of times as you become accustomed to breathing. Additionally, it can encourage breathing with your eyes closed. The brightness and color of the display can be changed and blinked so that you can breathe by closing your eyes and feeling the light in a dark place. For example, as shown in Figure 4e, the light can change from a dark black color to a red color that gradually becomes brighter as you breathe in. Also, as shown in Figure 4g, [Inhale (the display gradually changes from black at minimum brightness to red at maximum brightness) → Hold breath (display blinks) → Exhale (screen brightness and color gradually darken until it reaches minimum brightness). You can also set it to [Switch to] → Rest between sets (maintain display brightness and color)].
설정부(320)는, 사용자 단말(100)로부터 취침시각 및 기상시각을 설정받을 수 있다. 사용자 단말(100)은, 취침시각 및 기상시각을 설정할 수 있다. 도 4a와 같이 취침시각 및 기상시각에 대한 알람을 설정할 수 있는데, 12시간 형식 또는 24시간 형식의 시계를 그래픽 형태로 제공하고, 달 아이콘을 움직여 취침시각을 설정하도록 하며, 해 또는 알람시계 아이콘을 움직여 기상시각을 설정하도록 세팅할 수 있다. 또 도 4b와 같이 취침시각과 기상시각의 차이를 계산하여 예상수면시간을 알려줄 수도 있는데, 예상수면시간이 권장수면시간, 예를 들어 8 시간에 가까워질수록 시계모양 내 물이 채워지는 구조로 세팅될 수 있다. 이때, 사용자 단말(100)이 스마트폰일 수도 있지만, 상술한 바와 같이 스마트패드, 스마트워치, PC, TV, 헤드셋 등 디스플레이의 빛과 뇌파동조를 이용한 기상알람을 제공할 수 있다.The setting unit 320 may receive settings for bedtime and wake-up time from the user terminal 100 . The user terminal 100 can set bedtime and wake-up time. As shown in Figure 4a, you can set an alarm for bedtime and wake-up time. A 12-hour or 24-hour clock is provided in graphic form, and you can set the bedtime by moving the moon icon, and use the sun or alarm clock icon. You can set the wake-up time by moving it. In addition, as shown in Figure 4b, the difference between bedtime and wake-up time can be calculated to inform the expected sleep time. As the expected sleep time approaches the recommended sleep time, for example, 8 hours, water is filled in the clock shape. It can be. At this time, the user terminal 100 may be a smartphone, but as described above, it can provide a wake-up alarm using light and brain wave tuning from a display such as a smart pad, smart watch, PC, TV, or headset.
제어부(330)는, 사용자 단말(100)에서 설정한 취침시각 및 기상시각에 뇌파동조 및 자율신경계조절을 위한 빛, 사운드, 진동 및 호흡법 중 적어도 하나를 출력하도록 할 수 있다. 이때, 사용자 단말(100)은, 취침시각에 부교감신경계를 항진시키는 호흡법인 수면유도법을 출력하면서 멜라토닌 비억제 파장의 빛을 조사할 수 있다. 이때, 멜라토닌 비억제 파장의 빛은, 적색 및 황색의 빛을 포함할 수 있다. 그리고, 사용자 단말(100)은, 기상시각에 알파파 또는 베타파의 뇌파동조를 위하여 알파밴드(Alpha-Band) 또는 베타밴드(Beta-Band)의 플리커(Flicker)를 삽입하여 출력하고, 블루라이트를 출력하여 각성을 유도할 수 있다. 이때, 블루라이트는 애플리케이션 소프트웨어와 연동하는 LED 장치, 디스플레이, 플래시 라이트 등에 부착할 수 있는 블루라이트 밴드패스필터(Blue BanddPass Filter)를 이용하여 블루라이트를 이용한 각성을 유도할 수도 있다. 이때, 블루라이트가 각성을 촉진한다는 이론은, 마취상태에서 각성을 촉진시킨다는 논문(Liu D, Li J, Wu J, Dai J, Chen X, Huang Y, Zhang S, Tian B, Mei W. Monochromatic Blue Light Activates Suprachiasmatic Nucleus Neuronal Activity and Promotes Arousal in Mice Under Sevoflurane Anesthesia. Front Neural Circuits. 2020 Aug 18;14:55. doi: 10.3389/fncir.2020.00055. PMID: 32973462; PMCID: PMC7461971.)과, 수면중 블루라이트에 노출되면 DLMO(Dim Light Melatonin Onset)이 지연된다는 논문(Figueiro MG, Leggett S. Intermittent Light Exposures in Humans: A Case for Dual Entrainment in the Treatment of Alzheimer's Disease. Front Neurol. 2021 Mar 9;12:625698. doi: 10.3389/fneur.2021.625698. PMID: 33767659; PMCID: PMC7985540.)에 기반한다.The control unit 330 may output at least one of light, sound, vibration, and breathing methods for brain wave entrainment and autonomic nervous system control at the bedtime and wake-up time set in the user terminal 100. At this time, the user terminal 100 may irradiate light of a melatonin-inhibiting wavelength while outputting a sleep induction method, which is a breathing method that stimulates the parasympathetic nervous system at bedtime. At this time, light of the melatonin non-suppressing wavelength may include red and yellow light. In addition, the user terminal 100 inserts and outputs an alpha-band or beta-band flicker to synchronize alpha or beta brain waves at the time of waking up, and outputs blue light. Awakening can be induced by printing. At this time, blue light can also induce awakening using blue light using a blue light bandpass filter that can be attached to an LED device, display, or flash light that is linked to application software. At this time, the theory that blue light promotes awakening is based on the paper that it promotes awakening in anesthesia (Liu D, Li J, Wu J, Dai J, Chen X, Huang Y, Zhang S, Tian B, Mei W. Monochromatic Blue Light Activates Suprachiasmatic Nucleus Neuronal Activity and Promotes Arousal in Mice Under Sevoflurane Anesthesia. Front Neural Circuits. 2020 Aug 18;14:55. doi: 10.3389/fncir.2020.00055. PMID: 32973462; PMCID: PMC7461971.) and, blue light during sleep. A paper showing that exposure to DLMO (Dim Light Melatonin Onset) is delayed (Figueiro MG, Leggett S. Intermittent Light Exposures in Humans: A Case for Dual Entrainment in the Treatment of Alzheimer's Disease. Front Neurol. 2021 Mar 9;12:625698. doi: 10.3389/fneur.2021.625698. Based on PMID: 33767659; PMCID: PMC7985540.
기상뇌파유도부(340)는, 기상시각으로부터 기 설정된 시간 이전부터 사용자 단말(100)의 스피커를 통하여 세타파, 알파파 및 베타파의 뇌파동조를 위하여 모노럴비트(Monaural Beat) 또는 바이노럴비트(Binaural Beats)의 소리를 출력할 수 있다. 가장 낮은 델타파가 아닌 (낮은 주파수)세타-알파-베타(높은 주파수)를 유도하여 서서히 잠에서 깨고 각성할 수 있도록 세타파, 알파파 및 베타파의 뇌파동조를 순차적으로 유도할 수 있다. 이 뇌파동조의 이론은 상술한 뇌파동조이론과 같으므로 상세한 설명을 생략한다.The wake-up brain wave induction unit 340 generates a monaural beat or binaural beat for brain wave entrainment of theta waves, alpha waves, and beta waves through the speaker of the user terminal 100 from before a preset time from the wake-up time. Beats) sound can be output. By inducing theta-alpha-beta (higher frequency) rather than the lowest delta wave (lower frequency), brain wave entrainment of theta, alpha, and beta waves can be sequentially induced so that one can gradually wake up and wake up. Since this brain wave entrainment theory is the same as the brain wave entrainment theory described above, detailed explanation will be omitted.
수면단계별코칭부(350)는, 취침시각으로부터 사용자 단말(100)과 연동된 웨어러블 기기(400)에서 수면단계를 모니터링하고, 수면단계별로 기 설정된 모노럴비트 또는 바이노럴비트를 출력할 수 있다. 이때, 수면단계별코칭부(350)는 이어폰 또는 헤드폰이 사용자 단말(100)과 연동된 경우에는 바이노럴비트를 출력하고 스피커와 연결된 경우에는 모노럴비트를 출력할 수 있다.The sleep stage coaching unit 350 may monitor the sleep stage in the wearable device 400 linked to the user terminal 100 from bedtime and output a preset monaural beat or binaural beat for each sleep stage. At this time, the sleep stage coaching unit 350 may output binaural beats when earphones or headphones are linked to the user terminal 100 and output monaural beats when connected to a speaker.
인공지능부(360)는, 수면단계를 파악하기 위하여 웨어러블 기기(400)로부터 수집된 수집 데이터를 기 구축된 인공지능 알고리즘에 질의로 입력하여 수면단계를 파악할 수 있다. 수면단계는 웨어러블 기기(400)에서 수집된 GPS, 가속도센서, 지자기센서 등 얼마나 뒤척이고 뒤척이지 않는지, 심박수는 얼마인지 등에 따라 현재 수면단계의 어느 단계를 지나고 있는지를 파악하도록 할 수 있다.In order to determine the sleep stage, the artificial intelligence unit 360 can determine the sleep stage by inputting the collected data collected from the wearable device 400 as a query to a pre-built artificial intelligence algorithm. The sleep stage can be determined based on GPS, acceleration sensor, geomagnetic sensor, etc. collected by the wearable device 400, how much the person tosses and turns, what the heart rate is, etc.
<수면단계별 특징 및 생체신호센서><Characteristics and biological signal sensors for each sleep stage>
수면상태 판별을 위해서는 전체 수면 사이클 내의 수면단계에 대한 분석이 필수적이다. 일반적으로 인간의 수면 사이클은 WAKE에서 시작하여 크게 NREM 및 REM의 수면단계의 반복으로 이루어지며, 수면시 측정한 생체신호는 수면단계별로 서로 다른 특징을 가지게 된다. WAKE 단계에서는 근육이 활성화되어 있으며 호흡과 심장박동이 불규칙하다. NREM 단계에서는 심장박동, 호흡 및 눈의 움직임이 WAKE 단계에 비해 느려진다. NREM 수면단계 이후 REM 단계에 진입하게 되면, 심장박동과 호흡이 다시 빠르고 불규칙하게 변하며, 목 아래 근육의 긴장도가 저하된다.In order to determine sleep status, analysis of the sleep stages within the entire sleep cycle is essential. In general, the human sleep cycle starts from WAKE and largely consists of repetitions of the NREM and REM sleep stages, and biosignals measured during sleep have different characteristics for each sleep stage. In the WAKE stage, muscles are active and breathing and heartbeat are irregular. In the NREM stage, heart rate, breathing, and eye movements slow down compared to the WAKE stage. When you enter the REM stage after the NREM sleep stage, your heart rate and breathing become fast and irregular again, and the tension in the muscles below the neck decreases.
<특징 추출><Feature extraction>
<분당 호흡수><Respirations per minute>
호흡수는 사용자의 움직임에 영향을 받아 잡음이 발생한다. 따라서, 특정 지점에서 호흡수가 0이거나 값이 너무 크거나 작아 결측치라 판단되는 데이터는 앞뒤 값의 평균 데이터로 보정할 수 있다. REM 수면단계에서는 호흡이 불규칙한 특징을 보인다. 호흡이 불규칙해짐에 따라 진폭이 커지는 특성을 반영하여 크기가 3분인 윈도우의 최대 진폭을 측정하여 그 값을 특징으로 반영할 수 있다.Respiration rate is affected by the user's movements and noise is generated. Therefore, data that is judged to be missing because the respiratory rate is 0 or the value is too large or too small at a specific point can be corrected with the average data of the preceding and following values. In the REM sleep stage, breathing is characterized by irregularity. Reflecting the characteristic that the amplitude increases as breathing becomes irregular, the maximum amplitude of a window with a size of 3 minutes can be measured and that value can be reflected as a feature.
<분당 심박수><Heart beats per minute>
심장박동은 NREM 수면단계에 진입하는 순간 안정화되고, REM 수면단계에서는 심박수 그래프가 증가했다 감소하는 추이를 보이며, 다른 수면단계에 비해 진폭이 크게 나타난다. 이러한 특징을 반영하여 크기가 1분인 윈도우를 통해 진폭의 크기를 추출할 수 있다.The heart rate stabilizes the moment one enters the NREM sleep stage, and in the REM sleep stage, the heart rate graph shows an increasing and decreasing trend, and the amplitude appears larger than in other sleep stages. Reflecting these characteristics, the magnitude of the amplitude can be extracted through a window of 1 minute in size.
<심전도><Electrocardiogram>
심전도(ECG) 분석에 있어 보편적으로 사용되는 Pan-Tomkins 알고리즘에 따라QRS Complex를 추출할 수 있다. 심전도 데이터에서 추출할 수 있는 특징은 심박변이도(Heart Rate Variability, HRV), 진폭의 크기, 1 분 동안의 R 파(Wave) PEAK의 개수 총 3 종류가 있다. 심박 변이도의 분석은 QRS Complex에서 가장 큰 특징인 R-Peak 검출을 기반으로 이루어진다. 심박 변이도는 수면단계별로 다른 특징을 보이며,약 87% 이상의 정확도로 각성과 수면상태를 구분할 수 있다. 대역 통과 필터를 적용한 ECG 신호에서 수학식 1과 수학식 2를 활용하여 1 분 간격으로 R-R PEAK 간격의 표준편차를 구하고, 단기 심장박동 변이도 요소인 Short Term SDNN(Standard Deviation of NNInterval)을 구한다.QRS Complex can be extracted according to the Pan-Tomkins algorithm, which is commonly used in electrocardiogram (ECG) analysis. There are three types of features that can be extracted from ECG data: Heart Rate Variability (HRV), amplitude size, and number of R wave peaks per minute. Heart rate variability analysis is based on R-Peak detection, which is the biggest feature of the QRS Complex. Heart rate variability shows different characteristics depending on the sleep stage, and can distinguish between awakening and sleeping states with an accuracy of over 87%. Using Equation 1 and Equation 2 from the ECG signal to which a band-pass filter is applied, the standard deviation of the R-R PEAK interval is obtained at 1-minute intervals, and the Short Term SDNN (Standard Deviation of NNInterval), which is a short-term heart rate variability factor, is obtained.
<움직임><Movement>
3 축 가속도센서에서 수집된 데이터는 수면 중 움직임 여부를 판단하기 위해 사용된다. 방향성을 고려하지 않아도 되므로, 3 차원의 가속도 센서값을 1 차원의값으로 축소하여 수면중 움직임 강도를 표현한 Intensity 값으로 변환할 수 있다. XYZ 축 n 시점의 가속도를 이하 수학식 3에 대입하여 n 시점의 움직임 강도인 In을 얻는다.Data collected from the 3-axis acceleration sensor is used to determine movement during sleep. Since there is no need to consider directionality, the three-dimensional acceleration sensor value can be reduced to a one-dimensional value and converted into an intensity value expressing the intensity of movement during sleep. By substituting the acceleration at point n on the XYZ axis into Equation 3 below, In, the movement intensity at point n, is obtained.
<수면주기><Sleep cycle>
대부분의 생명체는 24시간 주기의 체내 시계(일주기리듬)를 가지고 있다. 측정된 생체신호에 상응하는 절대적인 시각 대신 수면주기를 나타내는 Clock Proxy라는 값을 부여하기 위해 기존에 사용되고 있는 생체시계 모델링 기법을 활용하여 측정시간 시작부터 끝나는 지점까지 값을 할당할 수 있다. X 축은 수면시간을 1초 간격으로 추출한 것이며, Y 축은 X 값에 대한 반주기의 사인(Sin) 그래프값을 할당한 것이다.Most living things have a 24-hour internal clock (circadian rhythm). In order to assign a value called Clock Proxy that represents the sleep cycle instead of the absolute time corresponding to the measured biological signal, the existing biological clock modeling technique can be used to assign a value from the start to the end of the measurement time. The
<데이터 생성, 학습 및 분류><Data creation, learning and classification>
센서별로 수집된 로우(Raw)데이터들은 수면단계별로 나타나는 파형을 판독하기 위해 시각화 한 뒤, 선행연구에서 관측되는 파형과 비교하여 WAKE 상태는 0, NREM 은 1, REM은 2의 레이블이 부여할 수 있다. Python Pandas와 Numpy 라이브러리를 활용하여 전처리 과정을 거친 뒤 측정시간을 주축으로 하나의 데이터 프레임으로 통합하여 CSV 파일로 저장할 수 있다. 해당 CSV 파일에는 심박변이도, 호흡변이도, 움직임강도, Clock Proxy가 특징으로 들어갈 수 있다. 전처리 과정을 마친 데이터들은 SVM 분류기의 학습 및 테스트 데이터로 이용할 수 있다.Raw data collected by each sensor can be visualized to read the waveforms that appear in each sleep stage, and then compared to the waveforms observed in previous studies, labels of 0 for WAKE state, 1 for NREM, and 2 for REM can be assigned. there is. After going through a preprocessing process using Python Pandas and the Numpy library, the measurement time can be integrated into one data frame and saved as a CSV file. The CSV file can include heart rate variability, breathing variability, movement intensity, and clock proxy as features. Data that has completed the preprocessing process can be used as training and testing data for the SVM classifier.
SVM 분류기는 각 클래스를 구분하는 결정 초평면(Decision Hyperplane)과 이것에 가장 인접한 샘플(Support Vector)의 거리를 최대하는 초평면을 구하는 것으로, 분류 문제에서 일반화 능력이 우수하다. 따라서 오류 데이터 영향이 적고 과적합이 되는 경우가 적다. Python Scikit-Learn 라이브러리를 활용하여 선형으로 구분될 수 없는 수면 데이터를 RBF(Radial Basis Function) 커널을 활용하여 초기 문제의 유한 차원보다 높은 차원으로 대응시켜 분류를 할 수 있다. 수면단계 분류의 성능을 높이기 위해 OvR(One-verses-Rest) 방식보다 처리 시간은 오래 걸리지만 OvO(One-verses-One) 방식을 선택하여 학습의 정확도를 높여줄 수 있다. RBF 커널 두 개의 매개변수인 C와 σ에 대해서는 Grid Search를 사용하여 경험적인 방법을 통해 각각 1과 0.1로 지정할 수 있다.The SVM classifier obtains a hyperplane that maximizes the distance between the decision hyperplane that separates each class and the closest sample (Support Vector), and has excellent generalization ability in classification problems. Therefore, the impact of error data is small and there are few cases of overfitting. Using the Python Scikit-Learn library, sleep data that cannot be linearly classified can be classified by matching them to a higher dimension than the finite dimension of the initial problem using the RBF (Radial Basis Function) kernel. To improve the performance of sleep stage classification, the accuracy of learning can be increased by selecting the OvO (One-verses-One) method, although it takes longer processing time than the OvR (One-verses-Rest) method. The two RBF kernel parameters, C and σ, can be specified as 1 and 0.1, respectively, through an empirical method using Grid Search.
수면 모니터링 데이터를 SVM 분류기에 넣어 수면단계를 분류한 결과를 바탕으로 총수면 대비 각 수면단계별 비율과 기록된 시간 중 실제 수면시간의 비율인 수면효율을 계산할 수 있다. 전체 수면시간 동안의 수면단계 변화 및 각 수면단계별 비율을 사용자에게 제공하기 위하여 Python Matplotlib 라이브러리를 활용하여 전체 수면시간에 대한 수면단계를 수면곡선(Hypnogram)과 수면단계별 비율을 파이차트로 시각화 할 수 있다. 수면곡선을 수면단계를 파악할 수 있으며 수면단계에 따른 취침준비 및 기상준비가 가능해질 수 있다.Based on the results of classifying sleep stages by putting sleep monitoring data into an SVM classifier, sleep efficiency, which is the ratio of each sleep stage to total sleep and the ratio of actual sleep time among the recorded time, can be calculated. In order to provide users with sleep stage changes during the entire sleep time and the ratio of each sleep stage, the Python Matplotlib library can be used to visualize the sleep stage for the entire sleep time as a sleep curve (Hypnogram) and the ratio of each sleep stage with a pie chart. . Sleep curves can be used to identify sleep stages, and it is possible to prepare for bed and wake up according to the sleep stage.
불면과다완화부(370)는, 사용자 단말(100)에서 불면증으로 등록하거나 불면증으로 등록하지 않은 경우, 수동설정으로 또는 수면단계별로 알파파, 세타파 및 델타파를 동조시키도록 모노럴비트 또는 바이노럴비트를 제공하고, REM 수면을 유도하기 위한 깊은 수면단계 후에 뇌파동조를 중지할 수 있다. 불면과다완화부(370)는, 사용자 단말(100)에서 수면과다증으로 등록하거나 수면과다증으로 등록하지 않은 경우, 수동설정으로 또는 세타파, 알파파 및 베타파의 순서로 동조시키도록 모노럴비트 또는 바이노럴비트를 제공할 수 있다. 여기서 불면증으로 등록하지 않은 경우란, 불면증으로 등록하지 않아도 가끔씩 잠이 안오는 경우도 존재하므로, 불면증은 아니지만 잠이 오지 않는 경우 등에 수동 또는 자동(수면단계)으로 뇌파동조를 유도하도록 하는 것을 의미한다. 수면과다증으로 등록하지 않은 경우란, 수면과다증으로 등록하지 않아도 가끔씩 잠이 너무 오는 경우도 존재하므로, 수면과다증은 아니지만 잠을 너무 자는 경우 등에 수동 또는 자동(수면단계)으로 뇌파동조를 유도하도록 하는 것을 의미한다.If the insomnia is registered as insomnia or not registered as insomnia in the user terminal 100, the excessive insomnia alleviation unit 370 uses monaural beats or binaural beats to synchronize alpha waves, theta waves, and delta waves by manual setting or by sleep stage. It provides beats and can stop brain wave entrainment after the deep sleep stage to induce REM sleep. If hypersomnia is registered in the user terminal 100 or not registered as hypersomnia, the insomnia hypersomnia alleviation unit 370 uses monaural beats or binaural beats to synchronize by manual setting or in the order of theta waves, alpha waves, and beta waves. Lalbit can be provided. Here, the case of not registering as insomnia means that brain wave entrainment is induced manually or automatically (sleep stage), such as in cases where insomnia is not registered but the patient cannot fall asleep, as there are cases where the patient cannot fall asleep even if the patient is not registered as insomnia. Cases that are not registered as hypersomnia mean that sometimes there are cases where you sleep too much even if you are not registered as hypersomnia. Therefore, in cases where you sleep too much but are not hypersomnia, it is recommended to induce brain wave synchronization manually or automatically (sleep stage). it means.
이때, 불면증의 경우 알파파, 세타파 및 델타파를 동조시키고, 수면과다증의 경우 세타파, 알파파 및 베타파의 순서로 동조시켜야 한다는 이론적 근거는 논문(Gantt MA. Study protocol to support the development of an all-night binaural beat frequency audio program to entrain sleep. Front Neurol. 2023 Jan 26;14:1024726. doi: 10.3389/fneur.2023.1024726. PMID: 36779067; PMCID: PMC9909225.)에 기반한다. 불면과다완화부(370)는, 이어폰 또는 헤드폰이 사용자 단말(100)과 연동된 경우에는 바이노럴비트를 출력하고 스피커와 연결된 경우에는 모노럴비트를 출력할 수 있다.At this time, the theoretical basis that in the case of insomnia, alpha waves, theta waves, and delta waves should be synchronized, and in the case of hypersomnia, the theta waves, alpha waves, and beta waves should be synchronized in that order, is provided in the paper (Gantt MA. Study protocol to support the development of an all) -night binaural beat frequency audio program to entrain sleep. Front Neurol. 2023 Jan 26;14:1024726. doi: 10.3389/fneur.2023.1024726. PMID: 36779067; PMCID: PMC9909225.) The excessive insomnia alleviation unit 370 may output binaural beats when earphones or headphones are linked to the user terminal 100 and output monaural beats when connected to a speaker.
이때, 나이가 증가하면 총 수면시간, 느린 주파수 수면(Slow Wave Sleep) 및 수면 효율이 감소하고, 수면개시 후 각성 및 수면지연시간이 증가할 수 있으므로 나이를 고려해야 하고, 여성은 일반적으로 남성보다 느린 주파수 수면을 유지하므로 성별도 고려할 수 있다. 상술한 불면증 또는 수면과다증이 완화되는 경우, 일주기리듬 장애, 양극성 장애, 우울증, 조울증, 불안장애, 알츠하이머병, 파킨슨병의 치료에 도움을 줄 수 있고, 여행자, 교대근무자 시차적응의 경우 일주기리듬을 재조정하는데 도움을 줄 수 있다.At this time, as age increases, total sleep time, slow frequency sleep (Slow Wave Sleep), and sleep efficiency decrease, and awakening and sleep delay time after sleep onset may increase, so age must be taken into consideration, and women generally have slower sleep than men. Since frequency sleep is maintained, gender can also be considered. When the above-mentioned insomnia or hypersomnia is alleviated, it can help treat circadian rhythm disorder, bipolar disorder, depression, manic depression, anxiety disorder, Alzheimer's disease, and Parkinson's disease, and in the case of jet lag for travelers and shift workers, the circadian rhythm disorder. It can help you re-establish your rhythm.
<수면골든타임 90분 법칙><90 minutes golden sleep time rule>
미국 스탠퍼드대 수면생체리듬(Sleep and Circadian Neurobiology) 연구소장 Seiji Nishino 박사의 베스트셀러 저서 "Stanford High-Efficiency Sleep Method (스탠퍼드식 최고의 수면법)"에서 소개한 바 있는 "The golden 90-minute sleep rule(수면 골든 타임 90분 법칙)"에 의하면, 도 4c와 같이 입면 직후 첫 번째 수면 사이클의 수면 퀄리티가 전체 수면 퀄리티에 중대한 영향을 미친다. 첫 수면주기에서 70 분 내지 90 분 동안 비렘수면 상태가 나타난다. 비렘수면 수면은 피로를 풀고 기억을 저장하는 깊은 잠의 단계다. 첫 비렘수면 수면이 이뤄지는 90분을 [수면의 골든타임]이라고 부른다. 잠든 직후 90분 동안 숙면하면 평소보다 적게 자더라도 다음 날 개운함을 느낄 수 있다는 것이다. 첫 주기 동안에 깊고 안정된 숙면을 취하게 된다면, 상대적으로 적은 시간 잘 수밖에 없는 현대인들에게 최선의 대안책이 될 수 있다. 우리 몸은 하루 중 깨어 있는 시간이 길어질수록 자고 싶은 욕구가 커진다. 이 욕구를 수면압력이라고 부른다. 수면압력은 잠이 든 후 90 분 동안, 즉 수면의 골든타임 때 가장 많이 방출된다. 수면 골든타임 때 숙면을 취하면 수면압력이 크게 줄어 자고 싶은 욕구가 해소되고 피곤함이 줄어든다. 수면 골든타임은 수면을 통해 자율신경을 조절하고, 성장호르몬 분비를 촉진시키며, 뇌 상태를 좋게 만든다."The golden 90-minute sleep rule" was introduced in the best-selling book "Stanford High-Efficiency Sleep Method" by Dr. Seiji Nishino, director of the Sleep and Circadian Neurobiology Laboratory at Stanford University. According to the "90-minute rule of time," as shown in Figure 4c, the sleep quality of the first sleep cycle immediately after waking up has a significant impact on the overall sleep quality. During the first sleep cycle, non-REM sleep occurs for 70 to 90 minutes. Non-REM sleep is a deep sleep stage that relieves fatigue and stores memories. The 90 minutes during which the first non-REM sleep occurs is called the [golden time of sleep]. If you sleep well for 90 minutes right after falling asleep, you can feel refreshed the next day even if you sleep less than usual. If you can get a deep and stable sleep during the first cycle, it can be the best alternative for modern people who have no choice but to sleep for a relatively short amount of time. The longer our bodies are awake during the day, the greater the desire to sleep. This desire is called sleep pressure. Sleep pressure is released the most during the 90 minutes after falling asleep, that is, during the golden time of sleep. If you get a good night's sleep during the golden sleep time, sleep pressure is greatly reduced, the desire to sleep is relieved, and fatigue is reduced. Sleep golden time regulates the autonomic nervous system through sleep, promotes the secretion of growth hormones, and improves brain condition.
여기서, 우울증 환자는 초기 비렘수면 시간이 부족하여 렘수면이 조기에 도래하기 때문에, 우울증을 치료하는 방법은 REM 수면 기상이다. 또, 논문(Palagini L, Baglioni C, Ciapparelli A, Gemignani A, Riemann D. REM sleep dysregulation in depression: state of the art. Sleep Med Rev. 2013 Oct;17(5):377-90. doi: 10.1016/j.smrv.2012.11.001. Epub 2013 Feb 5. PMID: 23391633.)에 따르면, REM 수면조절장애와 우울증은 강한 연관성이 존재한다. 이는 우울증의 치료 방법에 관계없이 발병, 재발, 재발의 위험과 심지어 치료 반응에 대한 영향까지 증가시키는 것으로 보인다. REM 수면의 핵심적인 역할은 우울증에서 정서적 반응성과 정서적 정보 처리를 조절하는 것이다. 우울증의 신경대사 변화는 REM 수면과잉 활성화에 의해 유발되거나 강화된다. 이에 따라, 본 발명의 일 실시예에서는, 델타파 뇌파 동조를 이용하여 첫 수면 사이클의 깊은 수면단계 시간을 증가시킬 수 있다. 수면단계를 모니터링하여 REM 수면 단계에 기상시켜 REM 수면과잉을 비활성화시킬 수 있다.Here, because patients with depression have insufficient initial NREM sleep time and REM sleep arrives early, the method of treating depression is REM sleep awakening. Also, a paper (Palagini L, Baglioni C, Ciapparelli A, Gemignani A, Riemann D. REM sleep dysregulation in depression: state of the art. Sleep Med Rev. 2013 Oct;17(5):377-90. doi: 10.1016/ According to j.smrv.2012.11.001. Epub 2013 Feb 5. PMID: 23391633.), there is a strong relationship between REM sleep regulation disorder and depression. This appears to increase the risk of onset, relapse, recurrence, and even impact on response to treatment, regardless of how depression is treated. A key role of REM sleep is to regulate emotional reactivity and emotional information processing in depression. Neurometabolic changes in depression are triggered or enhanced by REM sleep hyperactivation. Accordingly, in one embodiment of the present invention, the time of the deep sleep stage of the first sleep cycle can be increased using delta wave brain wave entrainment. By monitoring sleep stages, you can disable REM sleep hyperactivity by waking up in the REM sleep stage.
또한, REM instability에 의한 REM rebound(REM 지연 시간 단축 및 REM 밀도 증가)는 우울증 발생을 야기할 수 있다.Additionally, REM rebound (shortened REM latency and increased REM density) caused by REM instability can cause depression.
<비가청 주파수의 바이노럴비트><Binaural beats of inaudible frequencies>
비가청 주파수를 이용한 모노럴비트 또는 바이노럴비트를 통하여 뇌파를 동조시킬 수도 있다. 이는 논문(Choi MH, Jung JJ, Kim KB, Kim YJ, Lee JH, Kim HS, Yi JH, Kang OR, Kang YT, Chung SC. Effect of binaural beat in the inaudible band on EEG (STROBE). Medicine(Baltimore). 2022 Jul 1;101(26):e29819. doi: 10.1097/MD.0000000000029819. PMID: 35777013; PMCID: PMC9239629.)에 기반하는데, 일반적인 가청 주파수 바이노럴비트의 뇌파유도효과와 같이 비가청 주파수를 가진 바이노럴비트 역시 특정 뇌파의 유발이 가능하다는 점에서 출발한다.Brain waves can also be synchronized through monaural beats or binaural beats using inaudible frequencies. This is a paper (Choi MH, Jung JJ, Kim KB, Kim YJ, Lee JH, Kim HS, Yi JH, Kang OR, Kang YT, Chung SC. Effect of binaural beat in the inaudible band on EEG (STROBE). Medicine (Baltimore ). 2022 Jul 1;101(26):e29819. doi: 10.1097/MD.0000000000029819. PMID: 35777013; PMCID: PMC9239629.), which is based on non-audible frequencies, such as the EEG-induced effect of general audible frequency binaural beats. Binaural beats also start from the point that it is possible to induce specific brain waves.
햇빛샤워부(380)는, 기상시각에 맞추어 블루라이트인 청색의 빛을 제공할 수 있다. 이때, 햇빛샤워의 알람과 라이트테라피 시간을 도 4h와 같이 제공할 수 있다. 사용자 데이터를 통해 효과적인 사용시점 및 사용량을 알고리즘으로 예측하고 권장 사용시점과 사용량에 가까워질수록 시계 모양이 채워질 수 있다. 사용자의 수면패턴을 고려하여 최적화된 아침햇살샤워할 시간을 알려줄 수 있는데, 이때의 수면패턴은 언제 자고 언제 일어나는지이다. 또는, 언제 자야하고 언제 일어나야하는지에 따라 아침햇살샤워할 시간을 알려줄 수 있다. 불면증이나 수면과잉 등에 가장 좋은 것은 기상시각을 고정하는 것이고, 기상하자마자 햇볕을 보는 것이다. 블루라이트가 풍부한 아침햇살을 쬐면 오전에는 수면 호르몬인 멜라토닌이 억제되고, 세로토닌 호르몬이 분비되며 세로토닌 호르몬은 14 내지 15시간 후에 멜라토닌으로 변환되어 숙면을 유도한다. 설정하거나 예측되는 잠드는 시간보다 14 내지 15시간 전에 햇빛샤워를 하거나 라이트테라피를 하도록 시간을 안내하고 알람으로 알려줄 수 있다.The sunlight shower unit 380 can provide blue light, which is blue light, according to the waking time. At this time, the sunlight shower alarm and light therapy time can be provided as shown in Figure 4h. Effective usage timing and usage amount are predicted by an algorithm through user data, and the clock shape can be filled as the recommended usage timing and usage amount approaches. It can tell you the optimal time to take a morning shower by taking into account the user's sleep pattern. In this case, the sleep pattern is when you go to sleep and when you wake up. Or, it can tell you when to take a morning shower depending on when you should sleep and when you should wake up. The best thing for insomnia or oversleeping is to fix the waking time and see the sunlight as soon as you wake up. When exposed to morning sunlight rich in blue light, the sleep hormone melatonin is suppressed in the morning, the serotonin hormone is secreted, and the serotonin hormone is converted to melatonin after 14 to 15 hours to induce sound sleep. You can guide the time to take a sunlight shower or do light therapy 14 to 15 hours before the set or predicted bedtime and notify with an alarm.
일주기리듬맞춤부(390)는, 청색의 빛을 출력할 시간을 기상시각 이외에 사용자 단말(100)의 사용자 데이터로 파악된 일주기리듬에 기초하여 청색의 빛을 조사하도록 할 수 있다. 이때, 수면단계, 수면시간, 취침시각, 기상시각, 조도센서의 빛 노출시간, 빛 노출시각, GPS 데이터, 심박수 데이터 등 수면 데이터, 빛 노출 데이터, 활동 데이터를 측정할 수 있다. 그리고, 기 설정된 알고리즘을 이용하여 일주기리듬을 측정하고, 일주기리듬 장애 여부를 진단하며, 일주기리듬 데이터 분석을 통하여 개인 맞춤형 라이트테라피를 제공할 수 있다. 즉, 블루라이트의 노출시점, 노출강도 및 노출시간을 조절할 수 있다. 인간의 일주기시계는 24시간보다 약간 긴 주기로 실행되므로, 뇌파동조를 이용하여 매일 시간을 앞당길 수 있다.The circadian rhythm matching unit 390 may emit blue light based on the circadian rhythm determined by user data of the user terminal 100 in addition to the wake-up time. At this time, sleep data, light exposure data, and activity data such as sleep stage, sleep time, bedtime, wake-up time, light exposure time of the illuminance sensor, light exposure time, GPS data, and heart rate data can be measured. Additionally, circadian rhythm can be measured using a preset algorithm, circadian rhythm disorders can be diagnosed, and personalized light therapy can be provided through circadian rhythm data analysis. In other words, the exposure point, exposure intensity, and exposure time of blue light can be adjusted. Since the human circadian clock runs on a cycle slightly longer than 24 hours, brain wave entrainment can be used to advance the clock each day.
미주신경자극부(391)는, 일주기리듬의 장애가 모니터링된 경우, 미주신경자극기(NonInvasive Vagus Nerve Stimulation)에 대응하는 진동을 사용자 단말(100)에서 출력하도록 하거나 미주신경자극기와 연동시켜 미주신경자극기를 구동시킬 수 있다. 비침습 미주신경자극기는 고주파전기펄스(25 Hz, 2분)를 전달하는 휴대용 사용자 친화적 장치(gammaCore™)로, 미주신경의 구심성 섬유를 목의 피부에서 자극하고, 특히 삼차신경혈관 시스템으로 미주신경의 중앙 유입을 조절한다. 군발두통에 대한 nVNS의 효능이 두 개의 이중맹검 위약대조시험에서 평가되었는데, Headache(ACT-1) 연구는 133명의 환자에게 nVNS를 사용하여 5번의 군발두통발작을 치료하였고, 첫 번째 치료에서 위약 치료군과 비교하여 치료를 받은 전체 군발두통 환자군에서 치료 15분 내에 더 흔하게 통증이 경감되거나 사라지는 경향이 보고되었으며(26.7% vs. 15.1%, p=0.1), 특히 삽화군발두통 환자에서는 통계적으로 의미있는 차이가 보고되었다(34.2% vs. 10.6%, p=0.008). ACT-2 연구도 15분내 통증이 소실된 두통발작의 비율(48% vs.6%, p<0.01)로 nVNS 치료가 삽화군발두통군에서 위약 치료군보다 효과적인 연구 결과를 보였다. 이에 따라, 군발두통 환자군에서 중간 정도의 효능과 우수한 내약성이 입증되어, 표준요법을 받을 수 없거나 약물에 의해 충분한 완화를 보이지 않는 군발두통의 급성기 치료로 고려할 수 있다. 이에 대한 이론적 근거는 논문(Silberstein SD, Mechtler LL, Kudrow DB, Calhoun AH, McClure C, Saper JR, Liebler EJ, Rubenstein Engel E, Tepper SJ; ACT1 Study Group. Non-Invasive Vagus Nerve Stimulation for the ACute Treatment of Cluster Headache: Findings From the Randomized, Double-Blind, Sham-Controlled ACT1 Study. Headache. 2016 Sep;56(8):1317-32. doi: 10.1111/head.12896. PMID: 27593728; PMCID: PMC5113831.)에 근거한다.When a circadian rhythm disorder is monitored, the vagus nerve stimulation unit 391 outputs vibration corresponding to a vagus nerve stimulator (NonInvasive Vagus Nerve Stimulation) from the user terminal 100 or connects with the vagus nerve stimulator to generate a vagus nerve stimulator. can be driven. The non-invasive vagus nerve stimulator is a portable, user-friendly device (gammaCore™) that delivers high-frequency electrical pulses (25 Hz, 2 minutes) to stimulate the afferent fibers of the vagus nerve in the skin of the neck, and specifically to the trigeminal neurovascular system. Regulates the central inflow of nerves. The efficacy of nVNS for cluster headaches has been evaluated in two double-blind, placebo-controlled trials: the Headache (ACT-1) study, which treated five cluster headache attacks with nVNS in 133 patients, with no placebo treated in the first treatment. Compared to the overall group of cluster headache patients who received treatment, there was a tendency for pain to be more commonly relieved or disappear within 15 minutes of treatment (26.7% vs. 15.1%, p=0.1), and in particular, there was a statistically significant difference in episodic cluster headache patients. was reported (34.2% vs. 10.6%, p=0.008). The ACT-2 study also showed that nVNS treatment was more effective than the placebo treatment group in the episodic cluster headache group in terms of the proportion of headache attacks in which pain disappeared within 15 minutes (48% vs.6%, p<0.01). Accordingly, it has proven to have moderate efficacy and excellent tolerability in cluster headache patients, so it can be considered as an acute treatment for cluster headaches that cannot receive standard therapy or do not receive sufficient relief from drugs. The theoretical basis for this is in the paper (Silberstein SD, Mechtler LL, Kudrow DB, Calhoun AH, McClure C, Saper JR, Liebler EJ, Rubenstein Engel E, Tepper SJ; ACT1 Study Group. Non-Invasive Vagus Nerve Stimulation for the ACute Treatment of Cluster Headache: Findings From the Randomized, Double-Blind, Sham-Controlled ACT1 Study. Headache. 2016 Sep;56(8):1317-32. doi: 10.1111/head.12896. PMID: 27593728; PMCID: PMC5113831.) It is based on
이 외에도 Sensate®Somacoustics를 이용할 수도 있는데, 논문(Sensate® Somacoustics: A New Wave for Stress Management. Volume I By, Scott McDoniel, Ph.D., M.Ed.1 & Stefan Chmelik, M.Sc. 2 1. Faculty; College of Health Professions, Walden University, Minneapolis, MN 2. BioSelf Technologies, Ltd, London, UK)에 근거한다. 미국 FDA는 미주신경자극기(Vagus Nerve Stimulation)를 18세 이상 환자의 만성 재발성 우울증에 대한 장기 보조 치료제로 승인했다. 이때, 20 내지 30 Hz 사이의 전신 진동은 뇌의 분비 수준을 크게 증가시키고, 필요한 신경 가소성(Neural Plasticity)을 위한 단백질을 증가시킨다. 여기서, Sensate®는 스트레스 관리를 위한 비전기적인 진동성 장치인데, 이 장치는 저주파 기술(<50Hz)을 사용하여 만성 스트레스에 대한 부교감 신경계(PNS) 반응을 개선한다. Sensate® 장치의 두 가지 이론적 개념은 골전도(Bone Conduction)와 흉부 공명(Thoracic Resonance)이다. 흉골은 뼈의 움직임으로 인해 더 큰 증폭으로 이어질 수 있고 추가적인 에너지는 흉부 내 공명 특성으로부터 발생할 수 있다. 저주파 음파(60Hz)는 내부 장기를 중심으로 더 먼 거리까지 이동하고, 미주신경으로 이어지는 구심성 신경가지는 흉강 전체에 위치한다. 따라서, Sensate® 장치는 부교감신경계 기능을 항진시키는 새로운 미주 신경 자극 방법이다. 부교감신경계가 항진되면 불안, 불면증 및 기타 스트레스 관련 증상이 개선될 수 있다.In addition, you can also use Sensate®Somacoustics, which has a thesis (Sensate® Somacoustics: A New Wave for Stress Management. Volume I By, Scott McDoniel, Ph.D., M.Ed.1 & Stefan Chmelik, M.Sc. 2 1 Faculty; College of Health Professions, Walden University, Minneapolis, MN 2. BioSelf Technologies, Ltd, London, UK). The U.S. Food and Drug Administration (FDA) has approved Vagus Nerve Stimulation as a long-term adjunctive treatment for chronic recurrent depression in patients 18 years of age and older. At this time, whole body vibration between 20 and 30 Hz significantly increases the secretion level of the brain and increases proteins for necessary neural plasticity. Here, Sensate® is a non-electrical oscillatory device for stress management that uses low-frequency technology (<50 Hz) to improve parasympathetic nervous system (PNS) response to chronic stress. The two theoretical concepts of the Sensate® device are Bone Conduction and Thoracic Resonance. The sternum can lead to greater amplification due to bone movement and additional energy can come from resonance properties within the chest. Low-frequency sound waves (60 Hz) travel longer distances around internal organs, and the afferent nerve branches leading to the vagus nerve are located throughout the chest cavity. Therefore, the Sensate® device is a new vagus nerve stimulation method that enhances parasympathetic nervous system function. An increased parasympathetic nervous system can improve anxiety, insomnia, and other stress-related symptoms.
교대체감각자극부(393)는, 사용자 단말(100)과 연동된 웨어러블 기기(400)로부터 기 설정된 스트레스 지수를 초과하는 스트레스가 모니터링된 경우, 사용자 단말(100)을 사용자의 한 쪽 손 또는 손목에, 웨어러블 기기(400)를 사용자의 다른 쪽 손 또는 손목에 위치시키도록 한 후, 기 설정된 주파수를 가지는 진동을 출력하도록 할 수 있다. 이때, 진동의 주파수, 강도, 지속시간 및 횟수를 사용자 단말(100)에서 증감하도록 설정할 수 있다. 예를 들어, Wearable Stress Relief Device TouchPoints - TheTouchPoint Solution™(https://thetouchpointsolution.com/)에서 제공하는 터치포인트를 이용할 수 있는데, 터치포인트란 양 손목에 끼면 기 설정된 진동이 부드럽게 번갈아 전해지며 스트레스를 낮추는 손목밴드이다. 이는 논문(Leal-Junior, Ernesto Cesar Pinto et al. “A Triple-Blind, Placebo-Controlled Randomized Trial of the Effect of Bilateral Alternating Somatosensory Stimulation on Reducing Stress-Related Cortisol and Anxiety During and After the Trier Social Stress Test.” Journal of Biotechnology and Biomedical Science (2019): n. pag.)에 따르면, 교대체감각자극(Bi-Lateral Alternating Stimulation in Tactile)이 스트레스를 관리하는 비침습적이고 비약리학적인 수단을 제공할 수 있다고 한다. 즉, 양측의 교대적 신체감각자극이 주관적인 스트레스와 불안 수준을 줄이는 데 효과적일 수 있고, 불안을 가진 환자들에게 유익한 효과가 있다는 것이다.When stress exceeding a preset stress index is monitored by the wearable device 400 linked to the user terminal 100, the alternating somatosensory stimulation unit 393 moves the user terminal 100 to one hand or wrist of the user. After placing the wearable device 400 on the user's other hand or wrist, vibration with a preset frequency can be output. At this time, the frequency, intensity, duration, and number of vibrations can be set to increase or decrease in the user terminal 100. For example, you can use the touchpoints provided by Wearable Stress Relief Device TouchPoints - TheTouchPoint Solution™ (https://thetouchpointsolution.com/), which, when worn on both wrists, gently transmit preset vibrations alternately to relieve stress. It is a lowering wrist band. This is in the paper (Leal-Junior, Ernesto Cesar Pinto et al. “A Triple-Blind, Placebo-Controlled Randomized Trial of the Effect of Bilateral Alternating Somatosensory Stimulation on Reducing Stress-Related Cortisol and Anxiety During and After the Trier Social Stress Test.” According to the Journal of Biotechnology and Biomedical Science (2019): n. pag.), Bi-Lateral Alternating Stimulation in Tactile may provide a non-invasive, non-pharmacological means of managing stress. In other words, alternating bilateral physical sensory stimulation can be effective in reducing subjective stress and anxiety levels and has a beneficial effect on patients with anxiety.
또, 이 논문(Amy Serin, Nathan S. Hageman, Emily Kade (2018) The Therapeutic Effect of Bilateral Alternating Stimulation Tactile Form Technology on the Stress Response. Journal of Biotechnology and Biomedical Science - 1(2):42-47. https://doi.org/10.14302/issn.2576-6694.jbbs-18-1887)에서도 스트레스 반응을 매개하는 뇌 네트워크의 전기적 활동을 조절하여, 외상 후 스트레스 장애(PTSD)와 같이 불안수준이 높은 개인에게 스트레스 감소 효과를 제공한다고 기재되어 있다. 그리고, BLAST를 사용한 연구의 결과는 두 반구의 전기 활동 패턴이 빠르게 교대할 경우 뇌 반구 간 상호작용이 증가할 수 있다는 교대 반구 활성화 가설과 일치한다. Harper의 PTSD 대상자에 대한 뇌파 연구에 따르면 BLAST는 공포 기반 기억의 회상 동안 활성화되는 편도체의 시냅스에 대해 탈포텐시 효과가 있음을 발견했다. 이러한 결과는 BLAST가 스트레스와 불안과 관련된 주요 뇌 영역의 전기 활동에 영향을 미칠 수 있으며, 전반적인 영향은 인간의 스트레스 반응을 완화시키고 고통스러운 회상이나 신체적 고통과 관련된 신체 감각을 감소시키거나 제거할 수 있음을 시사한다. BLAST는 Salience 네트워크의 주요 영역의 전기적 활동을 줄임으로써 교감신경 활성화를 감소시킬 수 있다. 또, 강박장애를 가진 사람들은 주의력을 제한하고 내부의 산만함과 외부의 산만함으로 인해 과도한 경계를 형성하는 심각한 불안을 가질 수 있으므로, BLAST를 사용하여 과도한 활동이 크게 감소하는 것을 확인할 수도 있다.Also, this paper (Amy Serin, Nathan S. Hageman, Emily Kade (2018) The Therapeutic Effect of Bilateral Alternating Stimulation Tactile Form Technology on the Stress Response. Journal of Biotechnology and Biomedical Science - 1(2):42-47. https ://doi.org/10.14302/issn.2576-6694.jbbs-18-1887) also regulates the electrical activity of the brain network that mediates the stress response, helping individuals with high anxiety levels such as post-traumatic stress disorder (PTSD). It is said to provide a stress-reducing effect. Additionally, the results of studies using BLAST are consistent with the alternating hemisphere activation hypothesis, which suggests that rapid alternation of electrical activity patterns in the two hemispheres can increase interactions between brain hemispheres. Harper's electroencephalographic study of PTSD subjects found that BLAST has a depotentiating effect on synapses in the amygdala that are activated during recall of fear-based memories. These results suggest that BLAST may affect electrical activity in key brain regions associated with stress and anxiety, with the overall effect being to alleviate the stress response in humans and reduce or eliminate painful flashbacks or physical sensations associated with physical pain. It suggests that there is. BLAST can reduce sympathetic activation by reducing electrical activity in key regions of the salience network. Additionally, since people with OCD may have severe anxiety that limits their attention and creates hypervigilance to internal and external distractions, they may see a significant reduction in hyperactivity using BLAST.
<활용가능분야><Application areas>
논문(Roh HW, Choi JG, Kim NR, Choe YS, Choi JW, Cho SM, Seo SW, Park B, Hong CH, Yoon D, Son SJ, Kim EY. Associations of rest-activity patterns with amyloid burden, medial temporal lobe atrophy, and cognitive impairment. EBioMedicine. 2020 Aug;58:102881. doi: 10.1016/j.ebiom.2020.102881. Epub 2020 Jul 28. PMID: 32736306; PMCID: PMC7394758.)은 알츠하이머형 인지장애 환자가 비알츠하이머형 인지장애 환자보다 1시간 정도 더 늦게 깊은 잠에 들었다고 밝혔고, 논문(Kim J, Park I, Jang S, Choi M, Kim D, Sun W, Choe Y, Choi JW, Moon C, Park SH, Choe HK, Kim K. Pharmacological Rescue with SR8278, a Circadian Nuclear Receptor REV-ERBα Antagonist as a Therapy for Mood Disorders in Parkinson's Disease. Neurotherapeutics. 2022 Mar;19(2):592-607. doi: 10.1007/s13311-022-01215-w. Epub 2022 Mar 23. Erratum in: Neurotherapeutics. 2022 May 2;: PMID: 35322351; PMCID: PMC9226214.)에는 파킨슨병 환자들은 운동장애 뿐만 아니라 수면장애, 일주기리듬 장애를 겪으며, 그 중 하나가 저녁에 불안, 우울 등과 같은 정서장애를 보이는 일몰증후군이라고 한다. 또, 논문(이헌정. (2018). 일주기리듬의 조절이상이 양극성장애의 핵심발병기전일까?. 신경정신의학, 57(4), 276-286.)에서는 일주기리듬의 조절이상은 양극성장애와 밀접한 연관이 있고, 일주기리듬의 관리는 양극성장애의 예후 증진에 큰 도움이 되며 이를 위한 효과적인 치료 방법 개발이 필요하다고 기재되어 있다. 또, 아이메디신에 따르면, 경도인지장애(도 4i), ADHD(도 4j), 조현병(도 4k), 우울증(도 4l), 만성통증(도 4m), 불안장애(도 4n)의 뇌파를 정상인과 비교했는데 뇌파동조를 이용하여 정상에 가깝게 만들 수 있다면 상술한 질병 및 질환도 완화가 될 수 있다는 의미이다. 이에 따라, 경도인지장애, 알츠하이머치매, ADHD, 조현병, 우울증 및 만성통증에 대한 뇌파동조를 이용한다면 이 증상이나 질환이 완화될 수 있다.Paper (Roh HW, Choi JG, Kim NR, Choe YS, Choi JW, Cho SM, Seo SW, Park B, Hong CH, Yoon D, Son SJ, Kim EY. Associations of rest-activity patterns with amyloid burden, medial temporal lobe atrophy, and cognitive impairment. EBioMedicine. 2020 Aug;58:102881. doi: 10.1016/j.ebiom.2020.102881. Epub 2020 Jul 28. PMID: 32736306; PMCID: PMC7394758.) shows that patients with Alzheimer's type cognitive impairment have non-Alzheimer's type. They reported that they fell into a deep sleep about an hour later than patients with cognitive impairment, and the paper (Kim J, Park I, Jang S, Choi M, Kim D, Sun W, Choe Y, Choi JW, Moon C, Park SH, Choe HK, Kim K. Pharmacological Rescue with SR8278, a Circadian Nuclear Receptor REV-ERBα Antagonist as a Therapy for Mood Disorders in Parkinson's Disease. Neurotherapeutics. 2022 Mar;19(2):592-607. doi: 10.1007/s13311-022-01215- w. Epub 2022 Mar 23. Erratum in: Neurotherapeutics. 2022 May 2;: PMID: 35322351; PMCID: PMC9226214.), Parkinson's disease patients suffer from not only movement disorders but also sleep disorders and circadian rhythm disorders, one of which is evening disturbances. It is called sunset syndrome, which shows emotional disorders such as anxiety and depression. In addition, in the paper (Lee Heon-jeong. (2018). Is circadian rhythm dysregulation the core pathogenesis of bipolar disorder? Neuropsychiatry, 57(4), 276-286.), circadian rhythm dysregulation is bipolar disorder. It is closely related to circadian rhythm, and management of circadian rhythm is of great help in improving the prognosis of bipolar disorder, and it is necessary to develop effective treatment methods for this. In addition, according to iMediSync, the brain waves of mild cognitive impairment (Figure 4i), ADHD (Figure 4j), schizophrenia (Figure 4k), depression (Figure 4l), chronic pain (Figure 4m), and anxiety disorder (Figure 4n) Compared to normal people, this means that if brain wave entrainment can be used to bring the condition closer to normal, the diseases and conditions mentioned above can also be alleviated. Accordingly, if brain wave entrainment is used for mild cognitive impairment, Alzheimer's dementia, ADHD, schizophrenia, depression, and chronic pain, these symptoms or diseases can be alleviated.
| 질병disease | 빛light | 소리sound | 호흡Breath | 부교감신경Parasympathetic nervous system | 뇌파brain waves |
| 경도인지장애mild cognitive impairment | 감마밴드/알파팬드 플리커Gamma Band/Alpha Pand Flicker | 모노럴비트/바이노럴비트Monaural Beat/Binaural Beat | 깊은 호흡법deep breathing technique | 항진exaltation | 알파파alpha waves |
| ADHD/조현병/강박증ADHD/schizophrenia/obsessive-compulsive disorder | 블루라이트/베타밴드 플리커Blue light/beta band flicker | 모노럴비트/바이노럴비트Monaural Beat/Binaural Beat | 베타파beta waves | ||
| 학습력 향상Improved learning ability | 블루라이트/베타밴드 OR 감마밴드 플리커Blue light/beta band OR gamma band flicker | 모노럴비트/바이노럴비트Monaural Beat/Binaural Beat | 베타파(집중력)감마파(기억력)Beta waves (concentration) Gamma waves (memory) | ||
| 우울증depression | 블루라이트blue light | 모노럴비트(우뇌/왼쪽귀)Monaural beat (right brain/left ear) | 깊은 호흡법deep breathing technique | 항진exaltation | 알파파베타파alpha wave beta wave |
| 만성통증chronic pain | 블루라이트/알파밴드 플리커Blue light/alpha band flicker | 모노럴비트/바이노럴비트Monaural Beat/Binaural Beat | 깊은 호흡법deep breathing technique | 항진exaltation | 알파파alpha waves |
| 고혈압High blood pressure | 알파밴드 플리커alpha band flicker | 모노럴비트/바이노럴비트Monaural Beat/Binaural Beat | 깊은 호흡법deep breathing technique | 항진exaltation | 알파파alpha waves |
우울증의 경우 베타파를 활성화시키면서, 좌뇌와 우뇌의 뇌파 주파수 균형을 맞춰주는 방식의 뇌파동조를 시행할 수 있다. 목표는 알파파를 뇌의 왼쪽 반구 영역에서 줄이고 오른쪽 반구에서 증가시켜 균형을 맞추는 것이다. 이를 위해 우뇌의 알파파 동조를 위한 모노럴비트를 왼쪽 귀를 통해 들려준다. 블루라이트 테라피 시행 시 베타파 동조 모노럴비트 또는 바이노럴 비트를 재생한다. 깊은 호흡법을 코칭하여 부교감신경계를 항진시키고 알파파 뇌파를 강화시킨다. 또는 감마밴드의 모노럴비트 또는 바이노럴비트를 재생하여 감마 뇌파를 유도할 수도 있다. 그 이유는 우울증의 병인은 잘 알려져 있지 않지만 감마파 진동의 감소는 새로운 바이오마커라는 논문(Li Q, Takeuchi Y, Wang J, Gellert L, Barcsai L, Pedraza LK, Nagy AJ, Kozak G, Nakai S, Kato S, Kobayashi K, Ohsawa M, Horvath G, Kekesi G, L?rincz ML, Devinsky O, Buzsaki G, Berenyi A. Reinstating olfactory bulb-derived limbic gamma oscillations alleviates depression-like behavioral deficits in rodents. Neuron. 2023 Jul 5;111(13):2065-2075.e5. doi: 10.1016/j.neuron.2023.04.013. Epub 2023 May 9. PMID: 37164008; PMCID: PMC10321244.)이 존재하기 때문이다. 이때 도 4i 내지 도 4n의 뇌파(정량뇌파, QEEG)를 분석하면 이하 표 4와 같다. 여기서 도 4i 내지 도 4n의 출처는 상술한 바와 같이 아이메디신(iMediSync)이다. 이렇게 과도하거나 부족한 뇌파를 채워주는 동조를 실시함으로써 각 질환이나 증상이 완화될 수 있다.In the case of depression, brain wave entrainment can be performed by activating beta waves and balancing the brain wave frequencies of the left and right brains. The goal is to balance alpha waves by decreasing them in areas of the left hemisphere of the brain and increasing them in the right hemisphere. For this purpose, monaural beats to synchronize the alpha waves of the right brain are played through the left ear. When performing blue light therapy, play beta wave synchronized monaural beats or binaural beats. Coaching deep breathing techniques stimulates the parasympathetic nervous system and strengthens alpha brain waves. Alternatively, gamma brain waves can be induced by playing monaural beats or binaural beats of the gamma band. The reason is that although the etiology of depression is not well known, a decrease in gamma wave oscillations is a new biomarker (Li Q, Takeuchi Y, Wang J, Gellert L, Barcsai L, Pedraza LK, Nagy AJ, Kozak G, Nakai S, Kato S, Kobayashi K, Ohsawa M, Horvath G, Kekesi G, L?rincz ML, Devinsky O, Buzsaki G, Berenyi A. Reinstating olfactory bulb-derived limbic gamma oscillations alleviates depression-like behavioral deficits in rodents. Neuron. 2023 Jul 5;111(13):2065-2075.e5. doi: 10.1016/j.neuron.2023.04.013. Epub 2023 May 9. PMID: 37164008; PMCID: PMC10321244.) exists. At this time, analyzing the brain waves (quantitative brain waves, QEEG) of Figures 4i to 4n is shown in Table 4 below. Here, the source of FIGS. 4i to 4n is iMediSync, as described above. Each disease or symptom can be alleviated by entraining to compensate for excessive or insufficient brain waves.
| 도 4iFigure 4i | 경도인지장애mild cognitive impairment | -정상 노화 건강인의 뇌파에서는, 알파파와 같은 정상적인 뇌파가 관찰(노란색)-경도인지장애 환자의 뇌파에서는, 세타파와 같은 느린 뇌파가 관찰(초록색)-알츠하이머 치매 환자의 뇌파에서는, 델타파, 세타파와 같은 매우 느린 뇌파가 관찰(파란색)-In the EEG of normal aging healthy people, normal EEG waves such as alpha waves are observed (yellow) - In the EEG of patients with mild cognitive impairment, slow EEG waves such as theta waves are observed (green) - In the EEG of Alzheimer's dementia patients, delta waves and theta waves are observed. Very slow brain waves such as (blue) are observed (blue). |
| 도 4jFigure 4j | ADHDADHD |
-전두엽에서 알파파 과잉 + 알파 주파수는 정상인 패턴(회색)(-전두엽에서 알파파 과잉 + 알파 주파수가 느려지는 패턴(노란색)(-전두엽에서 세타파가 과잉되는 패턴(파란색)-베타 3파가 과잉되는 패턴(초록색)-Pattern with excessive alpha waves in the frontal lobe + normal alpha frequency (grey) (-pattern with excessive alpha waves in the frontal lobe + slowing down alpha frequency (yellow) (-pattern with excessive theta waves in the frontal lobe (blue) - |
| 도 4kdegree 4k | 조현병schizophrenia | - 전두엽에서 델타파가 과잉되는 패턴(파란색)-전두엽에서 세타파가 과잉되는 패턴(초록색)*이는 알파파 부족이 함께 나타날 수 있음- Pattern with excess delta waves in the frontal lobe (blue) - Pattern with excess theta waves in the frontal lobe (green) *This may appear together with lack of alpha waves. |
| 도 4lFigure 4l | 우울증depression | -전두엽에서 알파파가 과잉되는 패턴(초록색)-알파파가 느려지는 패턴(파란색)-전두엽에서 세타파가 과잉되는 패턴(노란색)-Pattern with excessive alpha waves in the frontal lobe (green) -Pattern with slow alpha waves (blue) -Pattern with excessive theta waves in the frontal lobe (yellow) |
| 도 4mdegree 4m | 만성통증chronic pain |
-두정엽에서 베타3파가 과잉되는 패턴(초록색)-Pattern in which |
| 도 4nFigure 4n | 불안장애anxiety disorder |
-중심엽에서 베타3파가 과잉되는 패턴(초록색)-알파파가 빨라지는 패턴(노란색)-Pattern with |
이하, 상술한 도 2의 관리 서비스 제공 서버의 구성에 따른 동작 과정을 도 3 및 도 4를 예로 들어 상세히 설명하기로 한다. 다만, 실시예는 본 발명의 다양한 실시예 중 어느 하나일 뿐, 이에 한정되지 않음은 자명하다 할 것이다.Hereinafter, the operation process according to the configuration of the management service providing server of FIG. 2 described above will be described in detail using FIGS. 3 and 4 as an example. However, it will be apparent that the embodiment is only one of various embodiments of the present invention and is not limited thereto.
도 3을 참조하면, (a) 관리 서비스 제공 서버(300)는 사용자 단말(100)과 웨어러블 기기(400)를 연동하여 저장하고, 사용자 단말(100)에서 취침시각 및 기상시각을 설정하는 경우, (b)와 같이 취침시각에는 수면유도를 실시하고 또 반대로 기상시각에는 각성을 뇌파동조 및 자율신경조절로 유도할 수 있다. 그리고 (c)와 같이 뇌파동조 및 자율신경조절 외에도 교대체감자극으로 스트레스 지수를 낮춰줄 수도 있고 (d)와 같이 미주신경자극으로 만성통증을 완화해줄 수도 있다. 그 외에도 다양한 뇌파동조로 증상을 완화할 수 있는 질병들은 표 3과 같으므로 중복된 설명은 생략하기로 한다. 또 도 4는 도 2에서 모두 설명했으므로 이 또한 중복된 설명은 생략하기로 한다.Referring to FIG. 3, (a) when the management service providing server 300 links and stores the user terminal 100 and the wearable device 400 and sets the bedtime and wake-up time in the user terminal 100, As shown in (b), sleep can be induced at bedtime, and conversely, awakening can be induced through brain wave entrainment and autonomic nerve control at wake-up time. In addition to brain wave synchronization and autonomic nerve control, as shown in (c), the stress index can be lowered through alternating sensory stimulation, and chronic pain can be alleviated through vagus nerve stimulation, as shown in (d). In addition, the diseases whose symptoms can be alleviated through various brain wave entrainment are listed in Table 3, so duplicate explanations will be omitted. In addition, since FIG. 4 has been fully described in FIG. 2, duplicate description thereof will also be omitted.
이와 같은 도 2 내지 도 4의 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 방법에 대해서 설명되지 아니한 사항은 앞서 도 1을 통해 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 방법에 대하여 설명된 내용과 동일하거나 설명된 내용으로부터 용이하게 유추 가능하므로 이하 설명을 생략하도록 한다.What is not explained about the method of providing real-time sleep health management service using AI-based brain wave entrainment and autonomic nervous system control in FIGS. 2 to 4 is the real-time sleep health management service using AI-based brain wave entrainment and autonomic nervous system control through FIG. 1. Since the management service provision method is the same as the description or can be easily inferred from the description, the description below will be omitted.
도 5는 본 발명의 일 실시예에 따른 도 1의 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템에 포함된 각 구성들 상호 간에 데이터가 송수신되는 과정을 나타낸 도면이다. 이하, 도 5를 통해 각 구성들 상호간에 데이터가 송수신되는 과정의 일 예를 설명할 것이나, 이와 같은 실시예로 본원이 한정 해석되는 것은 아니며, 앞서 설명한 다양한 실시예들에 따라 도 5에 도시된 데이터가 송수신되는 과정이 변경될 수 있음은 기술분야에 속하는 당업자에게 자명하다.FIG. 5 is a diagram illustrating a process in which data is transmitted and received between components included in the real-time sleep health management service providing system using AI-based brain wave tuning and autonomic nervous system control of FIG. 1 according to an embodiment of the present invention. Hereinafter, an example of the process of transmitting and receiving data between each component will be described with reference to FIG. 5, but the present application is not limited to this embodiment, and the process shown in FIG. 5 according to the various embodiments described above It is obvious to those skilled in the art that the process of transmitting and receiving data can be changed.
도 5를 참조하면, 관리 서비스 제공 서버는, 취침시각 및 기상시각에 뇌파동조 및 자율신경계조절을 위한 빛, 사운드, 진동 및 호흡법에 대한 설정을 저장할 수 있다.Referring to FIG. 5, the management service providing server may store settings for light, sound, vibration, and breathing methods for brain wave entrainment and autonomic nervous system control at bedtime and wake-up time.
그리고, 관리 서비스 제공 서버는, 사용자 단말로부터 취침시각 및 기상시각을 설정받고(S5200), 사용자 단말에서 설정한 취침시각 및 기상시각에 뇌파동조 및 자율신경계조절을 위한 빛, 사운드, 진동 및 호흡법 중 적어도 하나를 출력하도록 할 수 있다(S5300).In addition, the management service providing server receives the bedtime and wake-up time set from the user terminal (S5200), and selects light, sound, vibration and breathing methods for brain wave entrainment and autonomic nervous system control at the bedtime and wake-up time set by the user terminal. At least one can be output (S5300).
상술한 단계들(S5100~S5300)간의 순서는 예시일 뿐, 이에 한정되지 않는다. 즉, 상술한 단계들(S5100~S5300)간의 순서는 상호 변동될 수 있으며, 이중 일부 단계들은 동시에 실행되거나 삭제될 수도 있다.The sequence between the above-described steps (S5100 to S5300) is only an example and is not limited thereto. That is, the order between the above-described steps (S5100 to S5300) may change, and some of the steps may be executed simultaneously or deleted.
이와 같은 도 5의 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 방법에 대해서 설명되지 아니한 사항은 앞서 도 1 내지 도 4를 통해 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 방법에 대하여 설명된 내용과 동일하거나 설명된 내용으로부터 용이하게 유추 가능하므로 이하 설명을 생략하도록 한다.What is not explained about the method of providing real-time sleep health management service using AI-based brain wave entrainment and autonomic nervous system control in FIG. 5 is the real-time sleep health service using AI-based brain wave entrainment and autonomic nervous system control through FIGS. 1 to 4. Since the management service provision method is the same as the description or can be easily inferred from the description, the description below will be omitted.
도 7은 본 발명의 일 실시예에 따른 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템(11)의 구성을 나타낸 도면이다.Figure 7 is a diagram showing the configuration of a real-time sleep health management service providing system 11 using AI-based brain wave tuning and autonomic nervous system control according to an embodiment of the present invention.
이하에서는 본 발명의 일 실시예에 따른 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템(11)을 설명의 편의상 본 시스템(11)이라 하기로 한다.Hereinafter, the real-time sleep health management service providing system 11 using AI-based brain wave tuning and autonomic nervous system control according to an embodiment of the present invention will be referred to as the present system 11 for convenience of explanation.
도 7을 참조하면, 본 시스템(11)(즉, AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템)은 스마트 워치(1100) 및 분석 서버(1200)를 포함하여 구성될 수 있다.Referring to FIG. 7, this system 11 (i.e., a real-time sleep health management service providing system using AI-based brain wave entrainment and autonomic nervous system control) may be configured to include a smart watch 1100 and an analysis server 1200. there is.
본 시스템(11)은 스마트 워치(1100)를 이용해 수집(획득)되는 사용자의 생체 데이터를 이용하여 사용자의 수면을 코칭해주는 시스템을 의미한다. 본 시스템(11)은 스마트 워치(1100)를 통해 알 수 있는 사용자의 생체 데이터의 분석을 통해, 사용자의 수면의 질을 높일 수 있도록 하는 수면 코칭 기술을 제공할 수 있다. This system 11 refers to a system that coaches the user's sleep using the user's biometric data collected (obtained) using the smart watch 1100. This system 11 can provide sleep coaching technology to improve the user's sleep quality through analysis of the user's biometric data available through the smart watch 1100.
스마트 워치(1100)는 사용자 신체에 밀착되어 수면 중 사용자의 심박동, 움직임, 혈압을 포함하는 수면 생체 데이터를 수집할 수 있다. 본 발명의 실시예에서 사용자의 수면 생체 데이터를 수집하는 기기는 스마트 워치 등 다양한 웨어러블 기기를 포함할 수 있다. 즉, 본 발명에서 스마트 워치(1100)는 사용자의 신체의 일부에 착용되어 사용자의 수면 생체 데이터를 수집하는 웨어러블 기기를 의미할 수 있다. 본 발명에서 스마트 워치(1100)로는 스마트 워치 뿐만 아니라, 다양한 종류(형태)의 웨어러블 기기(무선통신 기기)가 적용될 수 있다. 스마트 워치(1100)는 수면 중인 상태(즉, 수면 중이어서 움직이지 않는 상태)에 있는 사용자의 수면 생체 데이터를 수집할 수 있다. 스마트 워치(1100)는 사용자가 소지한 웨어러블 기기를 의미할 수 있다.The smart watch 1100 is in close contact with the user's body and can collect sleep biometric data including the user's heart rate, movement, and blood pressure during sleep. In an embodiment of the present invention, devices that collect a user's sleep biometric data may include various wearable devices such as a smart watch. That is, in the present invention, the smart watch 1100 may refer to a wearable device that is worn on a part of the user's body and collects the user's sleep biometric data. In the present invention, the smart watch 1100 may be applied not only to a smart watch but also to various types (forms) of wearable devices (wireless communication devices). The smart watch 1100 may collect sleep biometric data of a user in a sleeping state (that is, in a state in which the user is sleeping and is not moving). The smart watch 1100 may refer to a wearable device owned by a user.
여기서, 수면 생체 데이터 내 움직임은, 일예로 스마트 워치(1100)의 움직임(모션, motion) 데이터에 대응되는 사용자의 신체의 움직임을 의미할 수 있고, 이는 스마트 워치(1100)에 내장된 모션 감지 센서 등을 이용해 획득되는 정보일 수 있다. 일예로 모션 감지 센서는 3축 가속도계, 3축 자이로스코프, 3축 자력계 등을 포함할 수 있다.Here, the movement in the sleep biometric data may mean, for example, the movement of the user's body corresponding to the movement (motion) data of the smart watch 1100, which is a motion detection sensor built into the smart watch 1100. This may be information obtained using, etc. For example, the motion detection sensor may include a 3-axis accelerometer, 3-axis gyroscope, 3-axis magnetometer, etc.
분석 서버(1200)는 수면 모니터링 알고리즘을 스마트 워치(1100)와 연동하여 사용자 개인에 맞춤화된 개인 맞춤형 수면 코칭에 대한 알람을 사용자에게 제공할 수 있다. 분석 서버(1200)는 일예로 인공지능을 이용한 실시간 수면 코칭 서버(장치) 등으로 달리 지칭될 수 있다.The analysis server 1200 may link a sleep monitoring algorithm with the smart watch 1100 to provide the user with an alarm for personalized sleep coaching tailored to each user. The analysis server 1200 may be referred to differently as, for example, a real-time sleep coaching server (device) using artificial intelligence.
분석 서버(1200)는, 인공지능을 이용한 실시간 수면 코칭과 관련된 웹 페이지, 앱 페이지, 프로그램 또는 애플리케이션을 분석 서버(1200)를 이용하는 사용자가 소지한 사용자 단말(미도시) 또는 스마트 워치(1100)로 제공할 수 있으며, 이를 통해 사용자가 개인 맞춤형 수면 코칭에 대한 알람을 제공받도록 할 수 있다. 여기서, 사용자 단말(미도시)은 사용자가 소지한 휴대 단말로서, 일예로 스마트폰(Smartphone), 스마트패드(SmartPad), 태블릿 PC, 노트북 등을 의미할 수 있다. 스마트 워치(1100)는 사용자 단말 중 한 종류로서, 웨어러블 디바이스를 의미할 수 있다.The analysis server 1200 provides web pages, app pages, programs, or applications related to real-time sleep coaching using artificial intelligence to a user terminal (not shown) or smart watch 1100 owned by a user using the analysis server 1200. This allows users to receive notifications for personalized sleep coaching. Here, the user terminal (not shown) is a portable terminal carried by the user and may mean, for example, a smartphone, smart pad, tablet PC, laptop, etc. The smart watch 1100 is a type of user terminal and may refer to a wearable device.
분석 서버(1200)는 스마트 워치(1100)로부터 사용자별 수면 생체 데이터를 수집하고 사용자의 움직임 여부에 따라 수면 생체 데이터를 축적할 수 있다. 일예로 분석 서버(1200)는 사용자가 움직이지 않는 동안의 심박수인 수면 생체 데이터를 축적할 수 있다. 또한, 분석 서버(1200)는 사용자가 움직이는 동안의 생체 데이터를 축정할 수 있다.The analysis server 1200 may collect biometric sleep data for each user from the smart watch 1100 and accumulate biometric sleep data depending on whether the user moves. For example, the analysis server 1200 may accumulate sleep biometric data, which is the heart rate while the user is not moving. Additionally, the analysis server 1200 can measure biometric data while the user moves.
분석 서버(1200)는 축적된 수면 생체 데이터를 인공지능 모델(즉, 딥러닝 모델)로 학습하고 분석함으로써 사용자별 수면 패턴 및 최적 수면주기 정보를 파악하고, 파악된 최적 수면주기 정보에 기반한 기상시간을 산출할 수 있다. 분석 서버(1200)는 기 저장된 수면주기 데이터와 최적 수면주기 정보를 사용자별 수면 생체 데이터를 통해 산출된 수면주기와 비교하여, 비교 결과에 따라 사용자의 수면주기를 예측하고 최적 수면주기 정보 및 최적 수면주기에 따른 기상시간을 산출할 수 있다.The analysis server 1200 learns and analyzes the accumulated sleep biometric data with an artificial intelligence model (i.e., deep learning model) to identify sleep patterns and optimal sleep cycle information for each user, and determines wake-up time based on the identified optimal sleep cycle information. can be calculated. The analysis server 1200 compares the previously stored sleep cycle data and optimal sleep cycle information with the sleep cycle calculated through sleep biometric data for each user, predicts the user's sleep cycle according to the comparison result, and provides optimal sleep cycle information and optimal sleep. Wake-up time can be calculated according to the cycle.
스마트 워치(1100)는 분석 서버(1200)로부터 기상시간 정보(즉, 분석 서버에서 산출된 기상시간의 정보인 기상시간 정보)를 전달받아, 기상시간에 웨이크업(wake up) 알람(즉, 기상알람, 기상알림)을 제공한다.The smart watch 1100 receives wake-up time information (i.e., wake-up time information, which is information about the wake-up time calculated by the analysis server) from the analysis server 1200, and sends a wake up alarm (i.e., wake-up time) at the wake-up time. Alarm, wake-up notification) are provided.
분석 서버(1200)는 스마트 워치(1100)와 같은 웨어러블 기기를 통해 사용자가 움직이지 않는 수면 중인 상태에서 수집한 사용자별 수면 생체 데이터를 축적할 수 있다. 이후, 분석 서버(1200)는 축적된 수면 생체 데이터를 인공지능 모델을 이용해 사용자별 수면 패턴을 분석하고, 사용자별 최적 수면주기 정보를 파악할 수 있다.The analysis server 1200 may accumulate sleep biometric data for each user collected while the user is sleeping and not moving through a wearable device such as a smart watch 1100. Afterwards, the analysis server 1200 can analyze sleep patterns for each user using the accumulated sleep biometric data using an artificial intelligence model and identify optimal sleep cycle information for each user.
본 발명에서, 사용자별 최적 수면주기 정보는 몇 번째 램 수면 시기에 기상했을 때 가장 개운한지에 대한 정보를 일반화한 것으로서, 분석 서버(1200)는 사용자별 최적 수면주기 정보를 기상시간에 따라 수집된 생체 정보의 분석을 통해 사용자 각각에 대해 파악할 수 있다.In the present invention, the optimal sleep cycle information for each user is a generalization of information about what REM sleep period the user feels most refreshed upon waking up, and the analysis server 1200 uses the optimal sleep cycle information for each user based on the biometric information collected according to the waking time. Through analysis of information, you can learn about each user.
본 발명에서 분석 서버(1200)가 스마트 워치(1100)로부터 수집하는 사용자의 생체 데이터(이는 생체정보라 달리 지칭될 수 있음)는, 일예로 i) 사용자가 수면 중 상태일 때(혹은 기상하지 않은 상태일 때) 측정된 생체 데이터(생체 정보)와 ii) 사용자가 수면에서 깬 상태인 활동 상태일 때(혹은 활동 중 상태일 때, 기상한 상태일 때) 측정된 생체 데이터(생체 정보)로 구분될 수 있다. 여기서, i) 경우에 해당하는 생체 데이터는 본 발명에서 수면 생체 데이터라는 용어로 달리 지칭될 수 있고, ii) 경우에 해당하는 생체 데이터는 본 발명에서 사용자 생체 정보(혹은 비수면 생체 데이터)라는 용어로 달리 지칭될 수 있다. 또한, 본 발명에서 생체 데이터(생체 정보, 생체 신호)는 심박수(심박동), 혈압, 움직임 등의 데이터(정보)를 의미할 수 있다. 또한, 본 발명에서 사용자의 상태는 크게 '수면 중 상태(즉, 수면 상태)'와 '수면이 아닌 상태'로 구분될 수 있고, 여기서, '수면이 아닌 상태'는 본 발명에서 수면 중이지 않은 상태(비수면 상태), 활동 상태, 수면에서 깬 상태, 기상상태(기상한 상태), 정상 상태 등으로 달리 지칭될 수 있다.In the present invention, the user's biometric data (which may be otherwise referred to as biometric information) collected by the analysis server 1200 from the smart watch 1100 is, for example, i) when the user is sleeping (or not waking up) It is divided into ii) biometric data (biometric information) measured when the user is awake from sleep (or active, awake). It can be. Here, the biometric data corresponding to case i) may be differently referred to in the present invention as sleep biometric data, and the biometric data corresponding to case ii) may be referred to by the term user biometric information (or non-sleep biometric data) in the present invention. It may be referred to differently. Additionally, in the present invention, biometric data (biometric information, biosignals) may mean data (information) such as heart rate (heart rate), blood pressure, and movement. In addition, in the present invention, the user's state can be largely divided into a 'sleeping state (i.e., a sleeping state)' and a 'non-sleeping state', where the 'non-sleeping state' refers to a state that is not sleeping in the present invention. It may be referred to differently as a state (non-sleep state), an active state, a waking state, a waking state (awake state), a normal state, etc.
분석 서버(1200)는 스마트 워치(1100) 등 웨어러블 기기를 이용해, 사용자의 움직임 유무를 분류하고, 움직임 없을 때의 심박수 데이터를 축적하고, 축적된 데이터 딥러닝을 통해 사용자의 수면단계를 예측 및 분류할 수 있다. 이후, 분석 서버(1200)는 수면단계, 심박수 변화량을 고려해 램(REM) 수면단계를 파악하고, 램(REM) 수면 이후 얕은 수면단계에서 스마트 워치(1100)를 통해 기상알람이 제공되도록 할 수 있다. 또한, 분석 서버(1200)는 사용자의 움직임과 심박수를 통해 사용자가 기상상태임을 확인할 때까지 알람 반복이 이루어지도록 할 수 있다. 즉, 분석 서버(1200)는 사용자가 기상상태임을 확인할 때까지 기상알람을 반복적으로 생성해 스마트 워치(1100)를 통해 사용자에게 반복 제공되도록 할 수 있다.The analysis server 1200 uses wearable devices such as a smart watch 1100 to classify the user's presence or absence of movement, accumulates heart rate data when there is no movement, and predicts and classifies the user's sleep stage through deep learning of the accumulated data. can do. Afterwards, the analysis server 1200 determines the REM sleep stage by considering the sleep stage and heart rate change, and provides a wake-up alarm through the smart watch 1100 in the light sleep stage after REM sleep. . Additionally, the analysis server 1200 may repeat the alarm until it confirms that the user is awake through the user's movements and heart rate. In other words, the analysis server 1200 can repeatedly generate a weather alarm and repeatedly provide it to the user through the smart watch 1100 until the user confirms that he or she is in a good weather state.
분석 서버(1200)는 스마트 워치(1100)를 통해 수집되는 생체 데이터(일예로 스마트 워치의 모션에 대응되는 사용자의 움직임, 심박수 등의 데이터)를 딥러닝 모델을 이용해 분석함으로써, 현재 사용자의 수면단계를 분류하고 예측할 수 있다.The analysis server 1200 analyzes biometric data collected through the smart watch 1100 (for example, data such as user movement and heart rate corresponding to the motion of the smart watch) using a deep learning model to determine the user's current sleep stage. can be classified and predicted.
분석 서버(1200)에서 고려되는 사용자의 수면단계는, 램(REM) 수면단계, 및 비램(논램) 수면단계를 포함할 수 있다. 비램 수면단계는 얕은 수면단계 및 깊은 수면단계를 포함할 수 있다. 램 수면단계는 깨어있는 상태(각성상태)와 얕은 수면단계의 사이에 위치하는 수면단계로서, 램수면단계에서 사용자는 꿈을 꾸게 될 수 있다. 얕은 수면단계는 램수면단계와 깊은 수면단계의 사이에 위치하는 수면단계로서, 일예로 얕은 잠이 이루어지는 1단계 수면단계 및 가벼운 수면이 이루어지는 2단계 수면단계를 포함할 수 있다. 깊은 수면단계는 얕은 수면단계보다 더 깊게 잠이 든 상태의 수면단계를 의미하는 것으로서, 일예로 서파가 나오는 수면 단계인 3단계 수면단계와 서파가 나오는 깊은 수면단계인 4단계 수면단계를 포함할 수 있다. 3단계 수면단계와 4단계 수면단계는 서파수면(slow-wave sleep) 단계라 지칭될 수 있다. 4단계 수면단계는 깨워도 일어나지 않고 업어가도 모르는 완전한 숙면상태를 의미할 수 있다.The user's sleep stage considered by the analysis server 1200 may include a REM sleep stage and a non-RAM sleep stage. The non-RAM sleep stage may include a light sleep stage and a deep sleep stage. The REM sleep stage is a sleep stage located between the waking state (awakening state) and the light sleep stage. In the REM sleep stage, the user may dream. The light sleep stage is a sleep stage located between the REM sleep stage and the deep sleep stage, and may include, for example, a stage 1 sleep stage in which light sleep occurs and a stage 2 sleep stage in which light sleep occurs. The deep sleep stage refers to a sleep stage in which you sleep more deeply than the light sleep stage. For example, it may include stage 3 sleep, a sleep stage in which slow waves appear, and stage 4 sleep, a deep sleep stage in which slow waves occur. there is. Stage 3 and stage 4 sleep can be referred to as slow-wave sleep stages. Stage 4 sleep can refer to a state of complete deep sleep in which the person does not wake up even if woken up or is not aware of being picked up.
또한, 분석 서버(1200)는 수면에 드는 시간이 평소패턴과 달라지면 전체 수면시간이 달라지게 되므로, 평소 패턴과 다른 시간에 수면을 시작한 경우, 사용자가 기상해야 하는 시간까지 남은 시간을 고려하여 1.5시간의 반복되는 수면주기에 적합한 최적의 기상시간(즉, 최적 기상시간)을 재조정하여 사용자를 깨울 수 있도록 한다. In addition, the analysis server 1200 determines that if the time taken to sleep is different from the usual pattern, the total sleep time will change, so if the user starts sleeping at a time different from the usual pattern, it takes into account the time remaining until the time the user has to wake up and sleeps for 1.5 hours. It adjusts the optimal wake-up time (i.e. optimal wake-up time) suitable for the user's repeated sleep cycle to wake the user.
또한, 분석 서버(1200)는 머신 러닝으로 학습한 사용자별 수면 생체 데이터를 통해 잠자리에 누워있는 시간 대비 실제 잠든 시간인 수면효율을 최적화해주는 취침시간(최적 취침시간)과 기상시간(최적 기상시간)을 산출하고 이를 취침알람 시간과 기상알람 시간으로 설정할 수 있다.In addition, the analysis server 1200 provides bedtime (optimal bedtime) and wake-up time (optimal wake-up time) that optimizes sleep efficiency, which is the actual sleeping time compared to the time lying in bed, through sleep biometric data for each user learned through machine learning. You can calculate and set this as the bedtime alarm time and wake-up alarm time.
또한, 분석 서버(1200)는 수면단계 판단 및 최적 주기 정보(최적 수면주기 정보) 파악에 최적화된 맞춤형 트레이닝 데이터 셋(Training Data Set)을 기반으로 딥러닝 뉴럴 네트워크에 기초한 기계학습을 수행하여 최적 기상시간 산출 모델을 구현할 수 있고, 구현된 최적 기상시간 산출 모델을 이용하여 최적 기상시간을 산출할 수 있다. In addition, the analysis server 1200 performs machine learning based on a deep learning neural network based on a customized training data set optimized for determining sleep stage and identifying optimal cycle information (optimal sleep cycle information) to determine optimal wake-up time. A time calculation model can be implemented, and the optimal wake-up time can be calculated using the implemented optimal wake-up time calculation model.
분석 서버(1200)는 스마트 워치(1100)와 네트워크를 통해 연동되어 데이터를 송수신할 수 있다.The analysis server 1200 can be linked with the smart watch 1100 through a network to transmit and receive data.
여기서, 네트워크는 일예로 3GPP(3rd Generation Partnership Project) 네트워크, LTE(Long Term Evolution) 네트워크, WIMAX(World Interoperability for Microwave Access) 네트워크, 인터넷(Internet), LAN(Local Area Network), Wireless LAN(Wireless Local Area Network), WAN(Wide Area Network), PAN(Personal Area Network), 블루투스(Bluetooth) 네트워크, NFC(Near Field Communication) 네트워크, 위성 방송 네트워크, 아날로그 방송 네트워크, DMB(Digital Multimedia Broadcasting) 네트워크 등을 포함할 수 있으나, 이에 한정된 것은 아니고, 다양한 유/무선 통신 네트워크를 포함할 수 있다.Here, the network includes, for example, 3rd Generation Partnership Project (3GPP) network, Long Term Evolution (LTE) network, World Interoperability for Microwave Access (WIMAX) network, Internet, Local Area Network (LAN), and Wireless Local (Wireless LAN). Area Network), WAN (Wide Area Network), PAN (Personal Area Network), Bluetooth network, NFC (Near Field Communication) network, satellite broadcasting network, analog broadcasting network, DMB (Digital Multimedia Broadcasting) network, etc. However, it is not limited to this and may include various wired/wireless communication networks.
도 8은 본 발명의 일 실시예에 따른 스마트 워치(1100)의 데이터 처리 구성을 나타낸 도면이다.Figure 8 is a diagram showing the data processing configuration of the smart watch 1100 according to an embodiment of the present invention.
도 8을 참조하면, 스마트 워치(1100)는 생체 데이터 수집모듈(1110), 통신모듈(1120), 수면 정보 제공 모듈(1130) 및 알람 모듈(1140)을 포함하여 구성될 수 있다.Referring to FIG. 8, the smart watch 1100 may be configured to include a biometric data collection module 1110, a communication module 1120, a sleep information provision module 1130, and an alarm module 1140.
본 명세서에서 사용되는 '모듈' 이라는 용어는 용어가 사용된 문맥에 따라서, 소프트웨어, 하드웨어 또는 그 조합을 포함할 수 있는 것으로 해석되어야 한다. 예를 들어, 소프트웨어는 기계어, 펌웨어(firmware), 임베디드코드(embedded code), 및 애플리케이션 소프트웨어일 수 있다. 또 다른 예로, 하드웨어는 회로, 프로세서, 컴퓨터, 집적 회로, 집적 회로 코어, 센서, 멤스(MEMS; MicroElectro-Mechanical System), 수동 디바이스, 또는 그 조합일 수 있다. 본 발명에서 '모듈'이라는 용어는 '부' 등으로 달리 지칭될 수도 있다.The term 'module' used in this specification should be interpreted to include software, hardware, or a combination thereof, depending on the context in which the term is used. For example, software may be machine language, firmware, embedded code, and application software. As another example, hardware may be a circuit, processor, computer, integrated circuit, integrated circuit core, sensor, MicroElectro-Mechanical System (MEMS), passive device, or a combination thereof. In the present invention, the term 'module' may also be referred to as 'unit', etc.
생체 데이터 수집모듈(1110)은 수면 중 사용자의 심박동, 움직임, 혈압을 포함하는 수면 생체 데이터를 수집하고, 사용자가 수면에서 깬 활동 상태일 때도 심박동, 움직임, 혈압을 포함하는 생체 데이터(즉, 사용자 생체 정보)를 수집할 수 있다. The biometric data collection module 1110 collects sleep biometric data including the user's heart rate, movement, and blood pressure while sleeping, and biometric data including heart rate, movement, and blood pressure even when the user is in an active state waking up from sleep (i.e., the user biometric information) can be collected.
통신모듈(1120)은 수집된 수면 생체 데이터와 생체 데이터(사용자 생체 정보)를 분석 서버(1200)로 전송하고, 분석 서버(1200)로부터 기상시간 정보를 수신한다. 통신모듈(1120)은 분석 서버(1200) 및 외부 서버와 데이터 통신하여 사용자의 최적 기상시간 및 수면 분석정보를 사용자에게 제공할 수 있도록 한다.The communication module 1120 transmits the collected sleep biometric data and biometric data (user biometric information) to the analysis server 1200, and receives wake-up time information from the analysis server 1200. The communication module 1120 communicates data with the analysis server 1200 and an external server to provide the user with the user's optimal waking time and sleep analysis information.
수면 정보 제공 모듈(1130)은 사용자에게 사용자의 평균 수면시간, 수면 시작시간, 기상시간을 포함하는 수면 패턴 정보를 제공하고, 매일의 수면시간 정보 및 부족한 수면시간 정보를 사용자에게 제공한다. 일예로, 수면 정보 제공 모듈(1130)은 그래프, 아이콘, 이미지 등의 시각적 객체를 이용해 사용자의 수면 패턴과 수면 패턴 변화 정보를 제공하여 사용자가 자신의 수면 정보를 시각적 객체를 통해 보다 직관적으로 이해할 수 있도록 한다. 수면 정보 제공 모듈(1130)은 이러한 다양한 정보들을 스마트 워치(1100)의 화면(디스플레이부)에 표시함으로써 사용자에게 제공할 수 있다. 또한, 수면 정보 제공 모듈(1130)은 수면시간이 부족한 경우(즉, 사용자의 수면시간이 부족한 것으로 판단되는 경우), 사용자가 낮잠, 쪽잠에 들 수 있는 시간을 안내하고, 사용자 피로도가 일정 수준인 경우, 이를 사용자에게 알릴 수 있다.The sleep information providing module 1130 provides the user with sleep pattern information including the user's average sleep time, sleep start time, and wake-up time, and provides the user with daily sleep time information and insufficient sleep time information. As an example, the sleep information providing module 1130 provides the user's sleep pattern and sleep pattern change information using visual objects such as graphs, icons, and images, so that the user can understand his/her sleep information more intuitively through visual objects. Let it happen. The sleep information providing module 1130 can provide such various information to the user by displaying it on the screen (display unit) of the smart watch 1100. In addition, when sleep time is insufficient (i.e., when it is determined that the user's sleep time is insufficient), the sleep information providing module 1130 guides the user on the time at which the user can take a nap or a short nap, and determines when the user's fatigue is at a certain level. In this case, this can be notified to the user.
알람 모듈(1140)은 분석 서버(1200)로부터 수신한 기상시간에 사용자 기상알람을 제공할 수 있다. 즉, 알람 모듈(1140)은 분석 서버(1200)로부터 수신한 기상시간 정보에 대응하는 시간에 기상알람을 제공할 수 있다. 여기서, 기상알람은 일예로 진동, 소리, 불빛 등으로 사용자에게 제공될 수 있다.The alarm module 1140 may provide a user wake-up alarm at the wake-up time received from the analysis server 1200. That is, the alarm module 1140 may provide a wake-up alarm at a time corresponding to the wake-up time information received from the analysis server 1200. Here, the wake-up alarm may be provided to the user in the form of vibration, sound, light, etc., for example.
도 9는 본 발명의 일 실시예에 따른 분석 서버(1200)의 데이터 처리 구성을 나타낸 도면이다.Figure 9 is a diagram showing the data processing configuration of the analysis server 1200 according to an embodiment of the present invention.
도 9를 참조하면, 분석 서버(1200)는 데이터 수집모듈(1210), 예측모듈(1220), 산출모듈(1230), 기상상태 확인 모듈(1240) 및 재 알람 생성 모듈(1250)을 포함하여 구성될 수 있다.Referring to FIG. 9, the analysis server 1200 includes a data collection module 1210, a prediction module 1220, a calculation module 1230, a weather condition confirmation module 1240, and a re-alarm generation module 1250. It can be.
데이터 수집모듈(1210)은 스마트 워치(1100)로부터 수면 중 사용자의 심박동, 움직임, 혈압을 포함하는 수면 생체 데이터를 수집하고, 수면 생체 데이터에 따라 사용자의 움직임 여부를 분류하고, 움직임이 없을 때의 사용자 심박수 데이터를 축적한다.The data collection module 1210 collects sleep biometric data including the user's heart rate, movement, and blood pressure during sleep from the smart watch 1100, classifies whether the user moves according to the sleep biometric data, and detects when there is no movement. Accumulate user heart rate data.
예측모듈(1220)은 축적된 데이터를 딥러닝을 통해 분석하여 사용자의 수면주기와 사용자의 생체 신호에 따른 실시간 수면단계를 예측한다. 예측모듈(1220)은 일예로 DNN(Deep Neural Network), CNN(Convolutional Neural Network), RNN(Recurrent Neural Network) 및 BRDNN(Bidirectional Recurrent Deep Neural Network) 중 적어도 하나를 포함하는 딥러닝 뉴럴 네트워크를 트레이닝 데이터 셋으로 학습시켜 수면단계 예측 모델을 구현할 수 있다. The prediction module 1220 analyzes the accumulated data through deep learning to predict the user's sleep cycle and real-time sleep stage according to the user's biological signals. For example, the prediction module 1220 uses a deep learning neural network including at least one of a Deep Neural Network (DNN), a Convolutional Neural Network (CNN), a Recurrent Neural Network (RNN), and a Bidirectional Recurrent Deep Neural Network (BRDNN) as training data. You can implement a sleep stage prediction model by learning in sets.
즉, 예측모듈(1220)은 기 학습된 딥러닝 모델을 이용하여 사용자의 수면단계를 예측할 수 있다. 본발명에서 고려되는 딥러닝 모델은 딥러닝, 인공지능(AI, artificial intelligence) 알고리즘 모델, 인공 신경망, 기계학습(머신러닝) 모델, 신경망 모델(인공 신경망 모델), 뉴로 퍼지 모델, 딥러닝 뉴럴 네트워크 등의 용어로 달리 지칭될 수 있다. 또한, 딥러닝 모델은 예시적으로 컨볼루션 신경망(Convolution Neural Network, CNN, 합성곱 신경망), 순환신경망(RNN, Recurrent Neural Network), 딥 신경망(Deep Neural Network) 등 종래에 이미 공지되었거나 향후 개발되는 다양한 신경망 모델이 적용될 수 있다. 본 발명에서 예측모듈(1220)이 사용자의 수면단계의 예측시 이용하는 딥러닝 모델은 기 학습된 딥러닝 모델로서, 기 학습된 수면단계 예측 모델 등의 용어로 달리 지칭될 수 있다.That is, the prediction module 1220 can predict the user's sleep stage using a previously learned deep learning model. Deep learning models considered in the present invention include deep learning, artificial intelligence (AI) algorithm model, artificial neural network, machine learning (machine learning) model, neural network model (artificial neural network model), neuro fuzzy model, and deep learning neural network. It may be referred to differently by terms such as: In addition, deep learning models include, for example, convolution neural networks (CNN), recurrent neural networks (RNN), and deep neural networks, which are already known or developed in the future. Various neural network models can be applied. In the present invention, the deep learning model used by the prediction module 1220 when predicting the user's sleep stage is a pre-learned deep learning model, and may be referred to differently by terms such as a pre-learned sleep stage prediction model.
산출모듈(1230)은 분석 서버(1200)에 미리 저장된 기상 시 가장 개운한 램 수면주기 정보를 기반으로 사용자별 최적 수면주기 정보를 산출한다. 일예로, 산출모듈(1230)은 산출된 최적 수면주기 정보에 따라 램(REM) 수면단계 이후 얕은 수면단계에서 사용자를 깨우는 기상시간을 산출할 수 있다.The calculation module 1230 calculates optimal sleep cycle information for each user based on the most refreshing RAM sleep cycle information upon waking up pre-stored in the analysis server 1200. For example, the calculation module 1230 may calculate the wake-up time to wake the user from the light sleep stage after the REM sleep stage according to the calculated optimal sleep cycle information.
또한, 산출모듈(1230)은 스마트 워치(1100)로부터 수신한 생체 정보(생체 데이터)의 분석을 통해, 사용자가 수면에 드는 수면 시작 시간을 파악하고, 수면 시작 시간을 평소 패턴과 비교할 수 있다. 이후, 산출모듈(1230)은 비교 결과에 따라 반복되는 수면주기를 고려하여 기상시간을 재조정할 수 있다. In addition, the calculation module 1230 can determine the sleep start time at which the user falls asleep through analysis of biometric information (biometric data) received from the smart watch 1100 and compare the sleep start time with the usual pattern. Thereafter, the calculation module 1230 may readjust the wake-up time by considering the repeated sleep cycle according to the comparison result.
예컨대, 평소 수면에 드는 시간보다 일찍 또는 늦게 잠들게 되면 수면시간이 달라지므로, 산출모듈(1230)은 기상해야 하는 시점까지 수면할 수 있는 남은 시간을 고려해 예측되는 마지막 수면주기 중 얕은 수면단계에 사용자가 일어날 수 있도록 기상시간을 산출할 수 있고, 이후, 산출된 기상시간을 스마트 워치(1100)로 전송하여 해당 시간(산출된 기상시간)에 기상알람을 제공받을 수 있도록 한다. For example, if you fall asleep earlier or later than your usual sleep time, your sleep time will change, so the calculation module 1230 considers the remaining time to sleep until you need to wake up and allows the user to sleep in the light sleep stage of the last sleep cycle predicted. The wake-up time can be calculated so that you can wake up, and then the calculated wake-up time is transmitted to the smart watch 1100 so that you can receive a wake-up alarm at the corresponding time (calculated wake-up time).
분석 서버(1200)는, 스마트 워치(1100)를 이용해 사용자의 움직임이나 심박수 등 생체 정보(생체 데이터)를 수집하고 분석하는 것뿐만 아니라, 딥러닝을 통해 사용자별 수면 습관 데이터를 획득한다. 여기서, 수면 습관 데이터는 사용자의 뒤척임 횟수, 수면 중 사용자가 중간에 깨는 횟수, 및 심박수 변화 정보를 포함할 수 있고, 이는 딥러닝 모델을 이용한 분석 및 학습을 통해 획득되는 데이터일 수 있다.The analysis server 1200 not only collects and analyzes biometric information (biometric data) such as the user's movements and heart rate using the smart watch 1100, but also acquires sleep habit data for each user through deep learning. Here, the sleep habit data may include the number of times the user tosses and turns, the number of times the user wakes up during sleep, and heart rate change information, and this may be data obtained through analysis and learning using a deep learning model.
일예로, 분석 서버(1200) 내 산출모듈(1230)은 획득한 수면 습관 데이터로부터 평균적으로 수면에 들어가는 실제 사용자의 수면 시작 시간과 비교하여 차이를 산출하고, 산출된 차이만큼의 시간을 계산해 마지막 수면주기 중 얕은 수면단계에 스마트 워치(1100)를 통해 사용자를 깨울 수 있도록 할 수 있다. 구체적인 예로, 산출모듈(1230)은, 사용자가 평소보다 1시간 늦게 잠들거나 일찍 잠에 든 경우, 기 정해진 시간에 일어나는 것이 아니라 일예로 도 6에 도시된 것과 같은 수면주기 그래프에 따라 1.5시간의 수면주기가 일정 횟수(일예로 3번 또는 4번) 반복된 이후의 시간까지만 자고 깨어날 수 있도록 기상시간을 산출할 수 있다. 예컨대, 산출모듈(1230)은 수면주기가 3번(3회) 반복된 4.5 시간 또는 4번(4회) 반복된 6시간 이후에 사용자가 기상하도록 기상시간(기상알람이 이루어지는 시간인 기상시간)을 산출할 수 있다.For example, the calculation module 1230 in the analysis server 1200 calculates the difference by comparing the sleep start time of the actual user who goes to sleep on average from the acquired sleep habit data, and calculates the time equal to the calculated difference to calculate the last sleep time. The smart watch 1100 can be used to wake the user during the light sleep phase of the cycle. As a specific example, if the user falls asleep 1 hour later or earlier than usual, the calculation module 1230 does not wake up at a predetermined time, but sleeps for 1.5 hours according to the sleep cycle graph as shown in FIG. 6, for example. The wake-up time can be calculated so that you can sleep and wake up only after the cycle is repeated a certain number of times (for example, 3 or 4 times). For example, the calculation module 1230 sets the wake-up time (wake-up time at which the wake-up alarm is made) so that the user wakes up after 4.5 hours when the sleep cycle is repeated 3 times (3 times) or 6 hours after the sleep cycle is repeated 4 times (4 times). can be calculated.
다시 말해, 분석 서버(1200)는, 일예로 스마트 워치(1100)로부터 수집된 사용자의 생체 데이터를 딥러닝 모델을 이용해 분석함으로써 사용자의 뒤척임 횟수나 중간에 깨는 횟수, 심박수 변화 등과 같은 사용자의 수면 습관 데이터를 도출할 수 있고, 이러한 사용자의 수면 습관 데이터 역시 딥러닝 모델로 학습할 수 있다. 분석 서버(1200)는 실시간으로 수집되는 사용자의 생체 데이터를 분석하여 매일마다 사용자의 움직임 변화와 심박수 변화를 도출할 수 있고, 이를 토대로 사용자가 현재 램 수면단계인지 여부를 파악하고, 사용자가 램 수면단계인 것으로 파악되면, 램 수면단계 이후의 얕은 수면단계에서 사용자가 기상할 수 있도록 산출되어 있는 기상시간에 맞추어 기상알람이 스마트 워치(1100)를 통해 제공되도록 할 수 있다. 즉, 분석 서버(1200)는 매일마다 실시간으로 사용자의 움직임과 심박수 변화 등을 분석하여, 설정한 기상시간 쯤에 이루어지는 얕은 수면단계에서(즉, 사용자의 수면단계가 램(REM) 수면단계 이후의 얕은 수면단계인 것으로 판단되는 시점에) 스마트 워치(1100)로 기상알람을 제공해 사용자를 깨워주는 기능을 제공할 수 있다.In other words, the analysis server 1200 analyzes the user's biometric data collected from the smart watch 1100 using a deep learning model to determine the user's sleep habits, such as the number of times the user tosses and turns, the number of times the user wakes up, and changes in heart rate. Data can be derived, and the user's sleep habit data can also be learned with a deep learning model. The analysis server 1200 can analyze the user's biometric data collected in real time to derive the user's movement changes and heart rate changes every day, and based on this, determine whether the user is currently in the REM sleep stage and determine whether the user is in the REM sleep stage. If it is determined that the user is in the light sleep stage after the REM sleep stage, a wake-up alarm can be provided through the smart watch 1100 according to the calculated wake-up time so that the user can wake up in the light sleep stage after the REM sleep stage. That is, the analysis server 1200 analyzes the user's movements and heart rate changes in real time every day, and determines whether the user's sleep stage is a light sleep stage that occurs around the set wake-up time (i.e., the user's sleep stage is after the REM sleep stage). A function that wakes the user up by providing a wake-up alarm to the smart watch 1100 (at a time when it is determined that the user is in a light sleep stage) can be provided.
또한, 사용자가 평소에 수면에 드는 시간보다 일찍 또는 늦게 잠들게 되면, 사용자의 당일 총 수면시간이 적어지거나 많아지게 되므로, 이에 분석 서버(1200)는, 매일마다 달라지는 사용자의 총 수면시간 중에서 기상해야 하는 시점(이는 분석 서버에 의해 산출된 기상시간을 의미함)까지 남아있는 남은 시간을 고려(즉, 사용자가 수면 중에 있는 현 시점으로부터 산출된 기상시간에 해당하는 시점까지 남아있는 남은 시간을 고려)하여 예측되는 마지막 수면 주기 중 얕은 수면단계에서 기상알람이 제공되도록 하여, 상기 얕은 수면단계에서 스마트 워치(1100)를 통해 제공되는 기상알람에 의해 사용자가 수면에서 깨도록 할 수 있다.In addition, if the user falls asleep earlier or later than the user's usual sleep time, the user's total sleep time for the day decreases or increases, so the analysis server 1200 determines which user's total sleep time should wake up, which varies every day. Considering the remaining time until the time point (this means the wake-up time calculated by the analysis server) (i.e., taking into account the remaining time remaining from the current point when the user is sleeping to the time corresponding to the calculated wake-up time) By providing a wake-up alarm in the light sleep stage of the predicted last sleep cycle, the user can wake up from sleep by the wake-up alarm provided through the smart watch 1100 in the light sleep stage.
또한, 산출모듈(1230)은 딥러닝 모델(머신러닝)으로 학습한 수면 생체 데이터를 통해 잠자리에 누워있는 시간 대비 실제 잠든 시간인 수면효율이 일정 수준 이상인(일정 수준 이상이 되는) 취침시간과 기상시간을 산출하여, 산출된 취침시간과 기상시간을 사용자에게 알림 하도록 할 수 있다. In addition, the calculation module 1230 calculates bedtime and wake-up time at which sleep efficiency, which is the actual sleeping time compared to the time lying in bed, is above a certain level (above a certain level) through sleep biometric data learned with a deep learning model (machine learning). By calculating the time, you can notify the user of the calculated bedtime and wake-up time.
다시 말해, 본 발명에서 고려되는 수면효율이라 함은 잠자리에 누워있는 시간 대비 실제 잠든 시간을 의미하는데, 분석 서버(1200)는 딥러닝 모델(머신러닝)로 학습한 사용자의 수면 생체 데이터를 이용하여 수면효율을 최적화해주는 취침시간과 기상시간을 산출(예측)하여 산출된 정보(즉, 최적 취침시간과 최적 기상시간)를 스마트 워치(1100)로 알림해 줄 수 있다. 분석 서버(1200)는 이러한 상기 산출된 정보를 사용자에게 제공함으로써, 사용자가 수면 효율이 최적이 되는 취침시간에 취침이 이루어지도록 하고, 수면 효율이 최적이 되는 기상시간에 기상할 수 있도록 하여 수면 질이 향상되도록 할 수 있다.In other words, sleep efficiency considered in the present invention means the actual sleeping time compared to the time lying in bed, and the analysis server 1200 uses the user's sleep biometric data learned with a deep learning model (machine learning) to Bedtime and wake-up time that optimize sleep efficiency can be calculated (predicted) and the calculated information (i.e., optimal bedtime and optimal wake-up time) can be notified to the smart watch (1100). The analysis server 1200 provides the calculated information to the user, allowing the user to go to bed at a bedtime when sleep efficiency is optimal and to wake up at a wake-up time when sleep efficiency is optimal, thereby improving sleep quality. This can be improved.
또한, 산출모듈(1230)은, 사용자의 수면 시작 시간이 평소 패턴(즉, 평소 수면 패턴)과 일정 수준 이상 달라지는 경우, 평소 패턴과 비교해 부족한 수면시간을 산출하여 알리고, 부족한 수면시간을 낮잠 또는 쪽잠으로 보충하도록 안내할 수 있다. In addition, if the user's sleep start time is different from the usual pattern (i.e., the usual sleep pattern) by a certain level or more, the calculation module 1230 calculates and notifies the insufficient sleep time compared to the usual pattern, and uses the insufficient sleep time to nap or take a nap. You can be guided to supplement with .
예컨대, 산출모듈(1230)은, 하루 동안의 필수 수면시간이 8시간(혹은 사용자의 수면 패턴을 분석하였더니, 일예로 사용자의 평소 수면시간이 평균 8시간)인데 일예로 실시간으로 수집된 사용자의 생체 데이터를 분석한 결과 사용자가 5시간 밖에 못잔 것으로 판단되는 경우, 사용자가 보충해야 하는 3시간 수면시간(즉, 3시간의 보충 수면시간)을 낮에 낮잠이나 쪽잠 방식으로 보충할 수 있도록 안내할 수 있다.For example, the calculation module 1230 determines that the required sleep time per day is 8 hours (or, upon analyzing the user's sleep pattern, the user's usual sleep time is 8 hours on average). For example, the user's sleep time collected in real time is 8 hours. If, as a result of analyzing biometric data, it is determined that the user has only slept 5 hours, the system will guide the user to make up for the 3 hours of sleep time (i.e. 3 hours of make-up sleep time) by taking a nap or napping during the day. You can.
달리 표현하면, 산출모듈(1230)은 일예로 스마트 워치(1100)로부터 실시간으로 수집된 사용자의 수면 생체 데이터를 분석하여, 사용자의 금일 수면이 이루어진 시간인 금일 수면시간을 도출하고, 이후 기 도출된 사용자의 필수 수면시간과 상기 도출된 금일 수면시간을 비교하여 금일 수면시간이 필수 수면시간보다 작은 것으로 판단되는 경우, 필수 수면시간과 금일 수면시간 간의 차이에 해당하는 시간을 보충 수면시간(혹은 부족한 수면시간)으로서 도출(산출)하고, 이후 도출된 보충 수면시간을 사용자가 낮잠 또는 쪽잠으로 보충할 수 있도록 안내하는 보충 수면시간 안내 정보를 생성하여 스마트 워치(1100)로 제공할 수 있다. 이러한 분석 서버(1200)는 보충 수면시간 안내 정보를 스마트 워치(1100)를 통해 사용자에게 제공함으로써, 사용자가 하루 수면시간의 총량(즉, 하루에 실제 수면이 이루어진 하루의 총 수면시간)을 매일매일 동일하게 유지 가능하도록 할 수 있다.In other words, the calculation module 1230 analyzes the user's sleep biometric data collected in real time from the smart watch 1100, derives today's sleep time, which is the time when the user slept today, and then calculates the previously derived sleep time. If today's sleep time is judged to be less than the required sleep time by comparing the user's required sleep time with today's sleep time, the time corresponding to the difference between the required sleep time and today's sleep time is supplemented with sleep time (or insufficient sleep) time), and then supplemental sleep time guidance information that guides the user to supplement the derived supplemental sleep time with a nap or a short nap can be generated and provided to the smart watch 1100. This analysis server 1200 provides supplementary sleep time guidance information to the user through the smart watch 1100, allowing the user to calculate the total amount of daily sleep time (i.e., the total sleep time of the day in which actual sleep was achieved) every day. It can be kept the same.
여기서, 보충 수면시간 안내 정보는 예시적으로 '평소보다(혹은 필수 수면시간보다) 3시간 수면이 부족한 상태이니, 낮에 3시간 동안의 낮잠을 주무시길 바랍니다' 등과 같을 수 있다. 또한, 상술한 설명에서 필수 수면시간은 일예로 일반적으로 사용자의 해당 나이에 필수적으로 요구되는 권장 수면시간으로 설정되거나, 또는 실시간으로 수집된 사용자의 생체 데이터의 분석을 기반으로 도출된 사용자의 평균 수면시간으로 설정될 수 있다.Here, the supplementary sleep time guidance information may, for example, be something like, 'You are lacking 3 hours of sleep compared to usual (or required sleep time), so please take a 3-hour nap during the day.' In addition, in the above description, the required sleep time is, for example, generally set to the recommended sleep time essential for the user's corresponding age, or the user's average sleep derived based on analysis of the user's biometric data collected in real time. Can be set to time.
또한, 분석 서버(1200)는 일예로 사용자가 낮잠을 취했을 때에도(낮잠을 취하고 있는 상태일 때에도) 스마트 워치(1100)를 이용해, 사용자의 수면 생체 데이터를 축적하여 분석하고, 수면주기를 고려해 사용자를 깨울 수 있도록 할 수 있다. 또한, 분석 서버(1200)는 사용자가 낮잠을 자고 있는 상태에서도 사용자가 완전히 일어났는지 여부를 생체 데이터의 분석을 통해 파악하여 반복적인 기상알람을 제공할 수 있도록 마련될 수 있다. 즉, 분석 서버(1200)는 스마트 워치(1100)를 통해 사용자에게 기상알람을 제공할 수 있고, 사용자의 움직임과 심박수가 기상상태임을 확인할 때까지 기상알람을 반복적으로 제공할 수 있다.In addition, for example, the analysis server 1200 uses the smart watch 1100 to accumulate and analyze the user's sleep biometric data even when the user takes a nap (even when the user is taking a nap), and analyzes the user by taking into account the sleep cycle. You can wake it up. Additionally, the analysis server 1200 may be configured to provide a repetitive wake-up alarm by determining whether the user has completely woken up even while the user is taking a nap through analysis of biometric data. That is, the analysis server 1200 can provide a wake-up alarm to the user through the smart watch 1100, and repeatedly provide the wake-up alarm until the user's movements and heart rate are confirmed to be in a wake-up state.
또한, 분석 서버(1200)는(특히, 분석 서버 내 산출모듈은) 사용자가 평소 낮잠을 자는 경우, 평소 수면에 드는 시간보다 일찍 또는 늦게 잠들게 되면 수면시간이 달라지므로, 기상해야 하는 시점까지의 남은 시간을 고려해 예측되는 마지막 수면주기 중 얕은 수면단계에 일어나도록 기상시간을 산출할 수 있다. In addition, the analysis server 1200 (in particular, the calculation module within the analysis server) determines that when the user usually takes a nap, the sleep time changes if the user falls asleep earlier or later than the usual sleep time, so the remaining time until the time to wake up is calculated. Considering the time, the wake-up time can be calculated to occur in the light sleep stage of the predicted last sleep cycle.
기상상태 확인 모듈(1240)은 스마트 워치(1100)로부터 사용자의 움직임, 심박수를 포함하는 생체 정보(생체 데이터)를 실시간으로 수신하고, 생체 정보를 분석하여 사용자의 생체 정보가 기상상태(기상한 상태)를 나타내는지 확인할 수 있다. 달리 말해, 기상상태 확인 모듈(1240)은, 실시간 수신되는 생체 정보가, 사용자가 수면에서 깬 활동 상태(혹은 기상한 상태, 정상 상태)일 때에 획득되는 사용자 생체 정보인지를 확인할 수 있다. 예컨대, 기상상태 확인 모듈(1240)은 기 저장된 수면에서 깬 정상 상태에서의 움직임 속도와 심박동수(심장 박동수)를 사용자의 스마트 워치(1100)에서 수집한 생체 데이터와 비교하여 사용자가 기상상태(기상한 상태)인지 여부를 파악할 수 있다. 일예로 기상상태 확인 모듈(1240)은, 사용자 움직임 속도가 기 저장된 정상 상태의 움직임 속도 이상이고, 심박동수가 기 저장된 정상 상태에서의 심박동수 이상인 경우, 사용자가 기상상태인 것(즉, 기상한 상태)으로 판단할 수 있다. The weather condition check module 1240 receives biometric information (biometric data) including the user's movements and heart rate from the smart watch 1100 in real time, analyzes the biometric information, and determines whether the user's biometric information is in a weather state (weather state). ) can be checked. In other words, the waking state confirmation module 1240 can check whether the biometric information received in real time is the user's biometric information obtained when the user is in an active state (or waking up state, normal state) after waking up from sleep. For example, the weather condition confirmation module 1240 compares the movement speed and heart rate (heart rate) in the normal state of awakening from previously stored sleep with biometric data collected from the user's smart watch 1100 to determine whether the user is in a normal state (weather condition). status) can be determined. For example, if the user's movement speed is higher than the pre-stored normal state movement speed and the heart rate is higher than the pre-stored heart rate in the normal state, the waking state confirmation module 1240 determines that the user is in a waking state (i.e., waking up state). ) can be judged.
재 알람 생성 모듈(1250)은 실시간으로 수신한 사용자의 생체 정보가 기상상태를 나타내지 않는 경우(즉, 실시간 수신한 사용자의 생체 데이터가 사용자 생체 정보가 아닌 것으로 판단되는 경우), 일정 시간 간격(일예로 5분 간격)으로 반복적으로 웨이크업 알람(기상알람)을 생성하고, 생성된 웨이크업 알람이 스마트 워치(1100)를 통해 제공되도록 할 수 있다. When the user's biometric information received in real time does not indicate the weather condition (i.e., when it is determined that the user's biometric data received in real time is not the user's biometric information), the re-alarm generation module 1250 is configured to operate at a certain time interval (e.g. A wake-up alarm (weather alarm) can be repeatedly generated at 5-minute intervals, and the generated wake-up alarm can be provided through the smart watch 1100.
이에 따르면, 분석 서버(1200)는 일예로 인공지능(딥러닝 모델)으로 사용자의 수면 생체 데이터 및 사용자 생체 정보(즉, 비수면 생체 데이터)를 포함한 생체 데이터(생체 정보)를 분석하는 과정을 통해 저장된 데이터(즉, 사용자에 대한 분석정보)와 사용자별 수면 생체 데이터를 비교함으로써, 사용자별로 수면단계를 예측하고 최적 주기 정보(최적 수면주기 정보) 및 기상시간을 산출할 수 있다.According to this, the analysis server 1200, for example, uses artificial intelligence (deep learning model) to analyze biometric data (biometric information) including the user's sleep biometric data and user biometric information (i.e., non-sleep biometric data). By comparing the stored data (i.e., user analysis information) with each user's sleep biometric data, it is possible to predict the sleep stage for each user and calculate optimal cycle information (optimal sleep cycle information) and wake-up time.
일예로, 분석 서버(1200)는 노이즈 대응 외 학습하지 못한 패턴 처리를 위해 학습 외 분포 데이터 탐지(out of distribution detection)과정을 수행한다. 학습 외 분포 데이터 탐지는 인공지능에 입력된 데이터가 학습된 확률분포 데이터 인지 아닌지 식별하는 것이다. 일예로, 분석 서버(1200)는 학습 외 분포 데이터 탐지를 통해 인공 신경망이 판단하기 어려운 이미지를 걸러내거나 예외 처리하여 안정성과 신뢰성을 높일 수 있도록 할 수 있다. 또한, 일예로 분석 서버(1200)는 학습 외 분포 데이터 탐지를 위해서 딥러닝 판정에 대해 얼마나 확신(confidence)하는지를 나타내는 확률 값을 보정(calibration)하거나 학습 외 분포 데이터를 생성적 대립 신경망(GAN, Generative Adversarial Network)으로 생성하고 학습하여 탐지 정확도를 향상시킬 수 있도록 할 수 있다. For example, the analysis server 1200 performs an out of distribution detection process other than learning to process unlearned patterns in addition to noise response. Non-learning distribution data detection is to identify whether the data input to artificial intelligence is learned probability distribution data. For example, the analysis server 1200 can improve stability and reliability by filtering out images that are difficult for the artificial neural network to judge or processing exceptions through detection of distribution data other than learning. In addition, as an example, the analysis server 1200 may calibrate a probability value indicating how confident it is in the deep learning decision in order to detect distribution data outside of learning, or use a generative adversarial neural network (GAN) to detect distribution data outside of learning. Adversarial Network) can be created and learned to improve detection accuracy.
또한, 분석 서버(1200)는 데이터 인식 정확도를 유지하면서 모델의 크기를 줄이기 위해, 연산을 간소화하는 경량 딥러닝 기술을 이용하여 사용자의 기상시간을 최종 확정할 수 있도록 한다. 일예로, 분석 서버(1200)는 데이터 인식을 위해 콘볼루션 신경망(CNN, Convolution Neural Network)에서 콘볼루션 필터를 변형하여 연산 차원을 축소(Reduction)하거나 큰 영향이 없는 신경망의 가중치(weight)를 삭제하는 가지치기, 가중치 값의 부동 소수점을 줄여 연산을 간소화하는 양자화 과정을 수행하여 데이터 경량화를 가능하도록 한다. 또한, 일예로 분석 서버(1200)는 미리 학습시킨 큰 신경망의 출력을 작은 신경망에서 모방 학습하도록 하여 연산을 간소화하며 정확도를 유지할 수 있도록 한다.In addition, the analysis server 1200 allows the user's wake-up time to be finalized using lightweight deep learning technology that simplifies calculations in order to reduce the size of the model while maintaining data recognition accuracy. For example, the analysis server 1200 transforms a convolutional filter in a convolutional neural network (CNN) to reduce the computational dimension or delete weights of the neural network that do not have a significant impact for data recognition. It performs pruning and quantization processes that simplify calculations by reducing the floating point number of weight values, enabling data lightweighting. In addition, as an example, the analysis server 1200 simplifies computation and maintains accuracy by imitating the output of a pre-trained large neural network and learning it from a small neural network.
또한, 분석 서버(1200)는 도면에 도시하지는 않았으나, 제어부(미도시)를 포함할 수 있다. 제어부(미도시)는 분석 서버(1200)에 포함된 각 모듈의 작동을 제어할 수 있고, 또한 스마트 워치(1100)의 작동(일예로 스마트 워치의 화면 표시)을 제어할 수 있다.Additionally, although not shown in the drawing, the analysis server 1200 may include a control unit (not shown). The control unit (not shown) can control the operation of each module included in the analysis server 1200, and can also control the operation of the smart watch 1100 (for example, displaying the screen of the smart watch).
분석 서버(1200)는 수면 시 소리를 이용해 수면을 모니터링하는 것이 아닌, 기 학습된 딥러닝 모델을 기반으로 사용자의 심박수와 움직임 등의 수면 생체 데이터를 이용해 사용자의 현재의 수면단계를 분류 및 예측할 수 있다. 분석 서버(1200)는 얕은 수면단계에서 사용자가 기상할 수 있도록 기상알람을 제공할 수 있는바, 이를 통해 사용자의 피로를 최소화하고 상쾌하게 기상할 수 있도록 도움 줄 수 있다.Rather than monitoring sleep using sound during sleep, the analysis server 1200 can classify and predict the user's current sleep stage using sleep biometric data such as the user's heart rate and movement based on a previously learned deep learning model. there is. The analysis server 1200 can provide a wake-up alarm to help the user wake up in a light sleep stage, thereby minimizing the user's fatigue and helping the user wake up refreshed.
분석 서버(1200)는 기 학습된 딥러닝 모델을 통해 사용자의 수면 상태에 맞추어 최적의 기상시간 및 취침시간을 찾아 그에 맞추어 깰 수 있도록 하는 기상알람을 제공 가능하되, 특히나 이를 애플 watch OS와 구글 wear OS 어플로 구현하여 스마트 워치(1100)를 통해 기상알람이 제공되도록 할 수 있다.The analysis server 1200 can find the optimal wake-up time and bedtime according to the user's sleep state through a pre-learned deep learning model and provide a wake-up alarm that allows the user to wake up accordingly, especially for Apple watch OS and Google wear. By implementing it as an OS application, a wake-up alarm can be provided through the smart watch (1100).
또한, 분석 서버(1200)에서 사용자의 수면단계의 분류 및 예측시 이용되는 기 학습된 딥러닝 모델은 본 발명에서 AI 알고리즘 등으로 달리 지칭될 수 있다. 이러한 기 학습된 딥러닝 모델(AI 알고리즘)은, 기존의 병원에서 진단하는 수면다원검사 결과와 비교했을 때 현재 사용자가 램(REM) 수면단계인지 여부를 100%의 확률로 동일하게 분류가 가능하도록 마련될 수 있다.In addition, the previously learned deep learning model used in the analysis server 1200 to classify and predict the user's sleep stage may be differently referred to as an AI algorithm or the like in the present invention. This pre-trained deep learning model (AI algorithm) is capable of equally classifying whether the user is currently in the REM sleep stage with a 100% probability when compared to the results of polysomnography diagnosed at existing hospitals. It can be provided.
분석 서버(1200)는 스마트 워치(1100)를 통해 수집된 사용자의 생체 데이터를 딥러닝 모델로 학습하여 사용자의 수면단계를 분류 및 예측할 수 있다.The analysis server 1200 can classify and predict the user's sleep stage by learning the user's biometric data collected through the smart watch 1100 using a deep learning model.
이때, 분석 서버(1200)에서 예측모듈(1220)은, 수면단계 예측을 위한 딥러닝 모델의 구현(생성)을 위해, 일예로 데이터 수집모듈(1210)에서 복수의 사용자가 소지한 복수개의 스마트 워치 각각으로부터 복수의 생체 데이터(수면 생체 데이터와 생체 정보가 포함된 생체 데이터)를 수집하면, 이후 i) 수집된 복수의 생체 데이터에 대한 전처리(preprocessing)를 수행하고, 이후 ii) 전처리된 복수의 생체 데이터에 대한 데이터 병합(merge)을 수행하며, 이후 iii) 병합된 데이터를 딥러닝 모델의 입력으로 적용하여 딥러닝 모델이 수면단계를 분류하도록 학습시킬 수 있다. 이에 대한 설명은 도 10 내지 도 14를 참조하여 보다 쉽게 이해될 수 있다.At this time, the prediction module 1220 in the analysis server 1200 uses, for example, a plurality of smart watches owned by a plurality of users in the data collection module 1210 to implement (generate) a deep learning model for predicting sleep stages. When a plurality of biometric data (sleep biometric data and biometric data containing biometric information) is collected from each, i) preprocessing is performed on the plurality of collected biometric data, and then ii) the plurality of preprocessed biometric data is processed. Data merging is performed on the data, and then iii) the merged data can be applied as input to the deep learning model to train the deep learning model to classify sleep stages. The explanation for this can be more easily understood by referring to FIGS. 10 to 14.
도 10 내지 도 14는 본 발명의 일 실시예에 따른 분석 서버에서 이용되는 딥러닝 모델을 설명하기 위한 도면이다.10 to 14 are diagrams for explaining a deep learning model used in an analysis server according to an embodiment of the present invention.
도 10 내지 도 14를 참조하면, 예측모듈(1220)은 전처리의 수행시, 사용자별로 수집된 심박수와 걸음수, 움직임 등의 데이터(생체 데이터)의 경우 시간이 모두 다르고 데이터의 샘플링 레이트(sampling rate)도 다 다르므로, 이를 고려하여 수집된 복수의 생체 데이터를 시간에 따라 교차 조인(cross join)을 하여 데이터를 구성하고, 이후 교차 조인하여 구성된 데이터에 대하여 스케일러(Scaler)를 통해 사용자(유저) 간의 차이를 없애고 불균형한 데이터를 제거함으로써 전처리된 복수의 생체 데이터를 마련할 수 있다. 일예로 예측모듈(1220)에 의해 전처리된 데이터(전처리된 복수의 생체 데이터)의 예는 도 10에 도시된 것과 같을 수 있다.Referring to FIGS. 10 to 14, when performing preprocessing, the prediction module 1220 monitors the data (biometric data) such as heart rate, number of steps, and movement collected for each user at different times and the sampling rate of the data. ) are all different, so considering this, the data is composed by cross-joining multiple collected biometric data according to time, and the data composed by cross-joining is then scaled to the user through a scaler. By eliminating the differences between the data and removing imbalanced data, multiple preprocessed biometric data can be prepared. For example, an example of data preprocessed by the prediction module 1220 (a plurality of preprocessed biometric data) may be as shown in FIG. 10.
여기서, 스케일러는 딥러닝 모델의 학습시 원본 데이터를 그대로 사용하게 되면, 원본 데이터의 경우 데이터 고유의 특성과 분포를 가지고 있음에 따라 학습이 느려지거나 문제가 발생하는 경우가 종종 존재하게 되므로, 이러한 문제의 해소를 위해 수집된 원본 데이터들(즉, 수집된 복수의 생체 데이터)를 서로 간에 동일하게 일정 범위로 스케일링하는 정규화하는 과정을 의미할 수 있다.Here, when the scaler uses the original data as is when training a deep learning model, learning often slows down or problems occur as the original data has unique characteristics and distribution, so these problems arise. It may refer to a process of normalizing the collected original data (i.e., a plurality of collected biometric data) by scaling them equally to a certain range.
전처리 수행 이후, 예측모듈(1220)은 데이터 병합(merge)의 수행시, 스케일러가 이루어진 전처리된 복수의 생체 데이터(즉, 모든 복수의 사용자의 생체 데이터)를 모두 합친 후(병합한 후), 레이블(Label)에 따른 불균형 데이터(imbalanced data) 처리를 수행할 수 있다. 일예로 병합된 데이터와 레이블과 심박수 간의 관계 그래프는 도 11에 도시된 것과 같을 수 있다.After performing preprocessing, when performing data merge, the prediction module 1220 combines (merges) all the preprocessed biometric data (i.e., biometric data of all multiple users) for which the scaler has been performed, and then labels them. Imbalanced data processing can be performed according to (Label). As an example, the relationship graph between merged data and labels and heart rate may be as shown in FIG. 11.
여기서, 불균형 데이터는 한 클래스(class)가 다른 클래스보다 숫자가 많은 것을 의미한다. 예측모듈(1220)은 불균형 데이터 처리의 수행시, 클래스 불균형을 위한 평가지표(일예로 ROC-AUC 평가지표)로 모델을 평가하는 방법, 불균형 데이터를 잘 처리하는 트리 기반 머신러닝 알고리즘(일예로 의사결정 트리, 랜덤 포레스트 등)을 활용하는 방법, 리샘플링(resampling)하는 방법 등을 이용할 수 있으며, 다만, 이에만 한정되는 것은 아니고, 종래에 기 공지되었거나 향후 개발되는 다양한 불균형 데이터 처리 기법이 적용될 수 있다.Here, imbalanced data means that one class has more numbers than the other class. When performing imbalanced data processing, the prediction module 1220 uses a method for evaluating the model using an evaluation index for class imbalance (e.g., ROC-AUC evaluation index), a tree-based machine learning algorithm that processes imbalanced data well (e.g., a doctor's Decision trees, random forests, etc.), resampling methods, etc. can be used, but the method is not limited to this, and various imbalanced data processing techniques known in the past or developed in the future can be applied. .
이후, 예측모듈(1220)은 병합된 데이터를 딥러닝 모델의 입력으로 적용하여 딥러닝 모델이 수면단계를 분류하도록 딥러닝 모델을 학습시킬 수 있는데, 이때 일예로 Xgboost와 같은 트리(tree) 기반의 기계학습(machine learning)으로 병합된 데이터들에 대해 수면단계 분류를 수행하고, 딥러닝 뉴럴 네트워크(Deep neural networks)를 통한 2D 분류(즉, 시간, 순서를 고려한 데이터 분류)를 수행할 수 있다. 이때, 예측모듈(1220)이 기계학습(머신러닝)을 이용한 수면단계 분류시 이용하는 오차행렬(Confusion matrix)은 일예로 도 12에 도시된 것과 같을 수 있고, 분류 결과는 도 14에 도시된 것과 같을 수 있다. 이에 따르면, 예측모듈(1220)은 움직임, 심박수 등의 생체 데이터를 통해 램 수면단계를 100%의 확률로 분류 가능한 기 학습된 딥러닝 모델을 구축하여, 이를 통해 사용자의 현재의 수면단계를 분류 및 예측할 수 있다. 또한, 분석 서버(1200)는 다음과 같은 기능들을 제공할 수도 있다. Afterwards, the prediction module 1220 can apply the merged data as an input to the deep learning model to train the deep learning model to classify the sleep stage. For example, a tree-based method such as Xgboost Sleep stage classification can be performed on data merged through machine learning, and 2D classification (i.e., data classification considering time and order) can be performed through deep learning neural networks. At this time, the error matrix (Confusion matrix) used by the prediction module 1220 when classifying sleep stages using machine learning (machine learning) may be, for example, as shown in FIG. 12, and the classification result may be as shown in FIG. 14. You can. According to this, the prediction module 1220 builds a previously learned deep learning model that can classify the REM sleep stage with 100% probability through biometric data such as movement and heart rate, and classifies and classifies the user's current sleep stage through this. It is predictable. Additionally, the analysis server 1200 may provide the following functions.
일예로, 분석 서버(1200) 내 기상상태 확인 모듈(1240)은 앞서 말한 바와 같이 스마트 워치(1100)가 기상알람을 제공하도록 한 후, 이후 사용자가 기상상태인지 여부를 확인하고, 이때 기상상태인 것으로 확인되면 지속적으로 스마트 워치(1100)를 통해 사용자의 생체 데이터를 수집(획득)하되, 만약 기상상태가 아닌 것으로 확인되면 재 알람 생성 모듈(1250)을 통해 스마트 워치(1100)에서 기상알람이 반복적으로 다시 제공되도록 할 수 있다. For example, the weather condition confirmation module 1240 in the analysis server 1200 causes the smart watch 1100 to provide a weather alarm as mentioned above, and then checks whether the user is in a weather condition, and at this time, determines whether the user is in a weather condition. If it is confirmed that the user's biometric data is continuously collected (obtained) through the smart watch (1100), but if it is confirmed that it is not a weather condition, the weather alarm is repeatedly issued by the smart watch (1100) through the re-alarm generation module (1250). It can be provided again.
이때, 기상상태 확인 모듈(1240)은, 사용자가 기상상태(즉, 기상한 상태)인 것으로 확인되면, 이후 인지 기능이 저하되어 있는 사용자의 수면 관성을 줄이기 위해 기상 후 운동을 목표 행동으로 설정하고, 이후 스마트 워치(1100)에서 상기 설정된 목표 행동이 인식될 때까지 기상알람이 기 설정된 주기(일예로 5분)마다 반복적으로 스마트 워치(1100)를 통해 사용자에게 제공되도록 할 수 있다.At this time, when it is confirmed that the user is in a waking state (i.e., waking up), the waking state confirmation module 1240 sets exercise after waking up as the target action to reduce the sleep inertia of the user with reduced cognitive function. , Afterwards, a wake-up alarm can be repeatedly provided to the user through the smart watch 1100 at preset intervals (for example, 5 minutes) until the set target behavior is recognized by the smart watch 1100.
달리 표현하면, 기상상태 확인 모듈(1240)은 사용자가 기상상태인 것으로 확인되면, 이후 기 설정되어 있는 목표 행동을 확인한 후 상기 기 설정되어 있는 목표 행동을 기반으로 스마트 워치(1100)에서 상기 확인된 목표 행동에 대응하는 생체 데이터가 획득될 때까지(즉, 실시간 수집되는 생체 데이터의 분석 결과 사용자가 목표 행동을 취한 것으로 감지될 때까지) 기상알람을 반복 생성하여 제공되도록 할 수 있다. 즉, 기상상태 확인 모듈(1240)은 사용자가 기상상태인 것으로 확인되었다 하더라도, 목표 행동에 대응하는 생체 데이터가 획득될 때까지, 기상알람이 기 설정된 주기마다 반복적으로(지속적으로) 제공되도록 할 수 있으며, 사용자가 목표 행동을 수행한 것으로 인식되면(즉, 목표 행동에 대응하는 생체 데이터가 획득된 것으로 감지되면), 기상알람이 더 이상 스마트 워치(1100)로 제공되지 않도록 기상알람의 제공을 종료(즉, 기상알람 제공을 OFF)시킬 수 있다.In other words, when the weather condition confirmation module 1240 confirms that the user is in a good weather condition, the weather condition confirmation module 1240 checks the preset target behavior and then performs the confirmed behavior in the smart watch 1100 based on the preset target behavior. A wake-up alarm can be generated and provided repeatedly until biometric data corresponding to the target behavior is obtained (that is, until it is detected that the user has taken the target behavior as a result of analysis of biometric data collected in real time). In other words, even if it is confirmed that the user is in a good weather state, the weather condition confirmation module 1240 can repeatedly (continuously) provide a wake-up alarm at preset intervals until biometric data corresponding to the target behavior is acquired. And, when the user is recognized as having performed the target action (i.e., when it is detected that biometric data corresponding to the target action has been obtained), the provision of the wake-up alarm is terminated so that the wake-up alarm is no longer provided to the smart watch 1100. (i.e., the provision of wake-up alarms can be turned OFF).
여기서, 목표 행동은 사용자가 기상한 후에 행동해야 하는 운동 정보를 의미하는 것으로서, 기상 후 운동, 기상 후 운동 미션 정보 등의 용어로 달리 지칭될 수 있다. 즉, 목표 행동은 사용자가 제공받은 기상알람을 끄기 위해 수행해야 하는 미션 정보를 의미할 수 있다. Here, the target action refers to exercise information that the user must take after waking up, and may be referred to differently by terms such as exercise after waking up or exercise mission information after waking up. In other words, the goal action may mean mission information that must be performed to turn off the wake-up alarm provided by the user.
이러한 목표 행동은 예시적으로 기상 후 10보(10걸음) 걷기, 3분간 걷기, QR/바코드 촬영하기, 스마트 워치(1100)와 사용자 단말(일예로 스마트폰 등) 간에 NFC 태깅 수행하기, 3분간 손 흔들기, 컵사진 찍기, 스쿼트 10번 하기, 기상 후 양치질 하기, 기상 후 줄넘기하기 등일 수 있으나, 이에만 한정되는 것은 아니고, 다양한 운동, 행동 등이 목표 행동으로 설정될 수 있다. 이러한 목표 행동의 설정은 분석 서버(1200)에 의해 자동으로 설정될 수도 있고, 또는 사용자에 의해 기 입력받음으로써 설정될 수도 있다.These target actions include, for example, walking 10 steps (10 steps) after waking up, walking for 3 minutes, shooting a QR/barcode, performing NFC tagging between the smart watch (1100) and a user terminal (e.g., a smartphone, etc.), and walking for 3 minutes. This may be waving your hand, taking a picture of a cup, doing 10 squats, brushing your teeth after waking up, or jumping rope after waking up, but it is not limited to these, and various exercises and actions can be set as target actions. This goal behavior may be set automatically by the analysis server 1200, or may be set by receiving input from the user.
또한, 앞서 말한 바와 같이, 분석 서버(1200)는 최적 수면주기에 따른 기상시간을 산출한 이후, 산출된 기상시간의 정보를 스마트 워치(1100)로 전달함으로써, 스마트 워치(1100)가 산출된 기상시간에 대응하는 시간에 기상알람(웨이크업 알람)을 사용자에게 제공하도록 스마트 워치(1100)의 작동을 제어할 수 있다.In addition, as mentioned above, the analysis server 1200 calculates the wake-up time according to the optimal sleep cycle and then transmits the information on the calculated wake-up time to the smart watch 1100, so that the smart watch 1100 calculates the wake-up time. The operation of the smart watch 1100 can be controlled to provide a wake-up alarm to the user at a time corresponding to the time.
이때, 분석 서버(1200)는 산출된 기상시간에 스마트 워치(1100)의 전원이 OFF 상태(방전된 상태)인 것으로 감지되는 경우, 산출된 기상시간의 정보를 사용자가 소지한 사용자 단말(미도시)로 제공함으로써, 사용자 단말에서 산출된 기상시간에 대응하는 시간에 기상알람(웨이크업 알람)이 울려 사용자에게 제공되도록 사용자 단말의 작동을 제어할 수 있다. 즉, 분석 서버(1200)는 스마트 워치(1100)의 사용 중에 전원이 방전되었을 때 알람 설정 시간(즉, 산출된 기상시간)에 사용자 단말에서 기상알람이 울리도록 하는 백업 알람 기능을 제공할 수 있다.At this time, when the analysis server 1200 detects that the power of the smart watch 1100 is in an OFF state (discharged state) at the calculated wake-up time, the analysis server 1200 sends the user terminal (not shown) holding information on the calculated wake-up time to the user. ), the operation of the user terminal can be controlled so that a wake-up alarm (wake-up alarm) is sounded and provided to the user at a time corresponding to the wake-up time calculated by the user terminal. In other words, the analysis server 1200 can provide a backup alarm function that causes a wake-up alarm to sound in the user terminal at the alarm setting time (i.e., calculated wake-up time) when the power is discharged while the smart watch 1100 is in use. .
즉, 분석 서버(1200)는, 기상시간에 스마트 워치(1100)에서 기상알람이 제공되도록 제어하되, 기상시간에 스마트 워치의 전원을 확인하여 스마트 워치의 전원이 OFF 상태(즉, 방전상태)인 것으로 감지되는 경우, 백업 알람 기능의 제공을 위해 사용자가 소지한 적어도 하나의 사용자 단말의 전원 ON/OFF 상태를 확인하여, 사용자가 소지한 적어도 하나의 사용자 단말 중 전원이 ON 상태에 있는 사용자 단말(즉, 전원이 ON 상태에 있는 것으로 확인된 사용자 단말)에서 기상알람이 제공되도록 제어할 수 있다. 여기서, 사용자가 소지한 적어도 하나의 사용자 단말에는 일예로 스마트폰, 태블릿 PC, 또 다른 스마트 워치 등 다양한 종류의 휴대 단말이 포함될 수 있다. 이러한 기능은 본 발명에서 백업 알람 기능이라 지칭될 수 있다. 분석 서버(1200)는 이러한 백업 알람 기능을 제공함으로써, 사용자가 소지한 복수의 사용자 단말들(일예로, 스마트 워치, 스마트폰, 태블릿 PC 등) 중에서 기상시간 정보를 전달받은 특정 사용자 단말이 방전상태임에 따라 기상알람의 제공이 불가능한 상태일 때, 다른 사용자 단말과 상호 보완적으로 방전상태가 아닌 사용자 단말에서(즉, 전원이 ON 상태로 켜져있는 사용자 단말)에서 기상알람이 울리도록 제어할 수 있다.That is, the analysis server 1200 controls the smart watch 1100 to provide a wake-up alarm at the wake-up time, but checks the power of the smart watch at the wake-up time to determine whether the smart watch is in an OFF state (i.e., in a discharge state). If it is detected, the power ON/OFF status of at least one user terminal owned by the user is checked to provide a backup alarm function, and at least one user terminal owned by the user whose power is in the ON state ( That is, it is possible to control the wake-up alarm to be provided from the user terminal (confirmed to be powered on). Here, at least one user terminal possessed by the user may include various types of portable terminals, such as a smartphone, a tablet PC, and another smart watch. This function may be referred to as a backup alarm function in the present invention. The analysis server 1200 provides this backup alarm function, so that a specific user terminal that has received wake-up time information among a plurality of user terminals (e.g., smart watch, smartphone, tablet PC, etc.) owned by the user is in a discharged state. Accordingly, when it is impossible to provide a wake-up alarm, the wake-up alarm can be controlled to sound in a user terminal that is not in a discharged state (i.e., a user terminal that is turned on) in a complementary manner with other user terminals. there is.
또한, 분석 서버(1200)는 기상알람이 사용자에게 제공되도록 할 때, 일예로 사용자의 마음이 진정되는 조용한 자연음 소리가 스마트 워치(1100)를 통해 출력되도록 하는 1단계, 스마트 워치(1100)에 진동이 울리게 하는 2단계, 스마트 워치(1100)에서 알람 벨소리가 출력되도록 하는 3단계 및 사용자 단말에서 알람 벨소리가 출력되도록 하는 4단계의 순서로 기상알람이 제공되도록 할 수 있다.In addition, when the analysis server 1200 provides a wake-up alarm to the user, for example, the first step of outputting a quiet natural sound that soothes the user's mind through the smart watch 1100 is to the smart watch 1100. A wake-up alarm can be provided in the following order: step 2 of causing vibration, step 3 of outputting an alarm ringtone from the smart watch 1100, and step 4 of outputting an alarm ringtone from the user terminal.
이때, 분석 서버(1200)는 스마트 워치(1100)를 통해 기상알람이 제공되고 있는 상태일 때(즉, 스마트 워치(1100)에서 기상알람이 울리고 있는 상태일 때), 일예로 사용자 단말이 스마트 워치(1100)와 연동(연결) 가능한 근거리 통신거리(일예로 NFC, 블루투스 통신거리) 이내에 위치하는 것으로 감지되면서, 사용자 단말이 사용자의 손에 파지된 상태인 것으로 감지되는 경우에 한하여, 제공되는 기상알람이 해제(즉, 제공되는 기상알람이 OFF되도록) 제어할 수 있다. 달리 표현하면, 분석 서버(1200)는 일예로 기상알람이 제공되고 있는 상태일 때, 사용자가 자신이 소지한 사용자 단말(일예로 스마트폰)을 스마트 워치(1100)와 연동 가능한 원거리(근거리) 장소에 위치시키고 사용자가 사용자 단말에 도달해야만 기상알람이 해제(OFF)되도록 제어할 수 있다. 다만, 이에만 한정되는 것은 아니고, 분석 서버(1200)는 일예로 사용자가 기상한 것을 스마트 워치(1100)의 AI 알고리즘이 감지하거나, 혹은 분석 서버(1200)가 실시간 수집되는 생체 데이터의 분석을 통해 기상상태인 것으로 감지하게 되면, 이후 자동으로 기상알람이 해제(정지, OFF) 되도록 제어할 수 있다.At this time, when the analysis server 1200 is in a state where a wake-up alarm is being provided through the smart watch 1100 (i.e., when the wake-up alarm is ringing in the smart watch 1100), for example, the user terminal is connected to the smart watch. A weather alarm is provided only when the user terminal is detected to be held in the user's hand while being located within a short-distance communication distance (e.g., NFC, Bluetooth communication distance) that can be linked (connected) with (1100). This can be controlled to turn off (i.e., turn off the provided wake-up alarm). In other words, the analysis server 1200 is a remote (near-distance) location where the user can link his/her user terminal (e.g. a smartphone) with the smart watch 1100 when a weather alarm is being provided. It can be controlled so that the wake-up alarm is turned off only when the user reaches the user terminal. However, it is not limited to this, and the analysis server 1200 detects that the user wakes up, for example, through the AI algorithm of the smart watch 1100, or through the analysis of biometric data collected in real time by the analysis server 1200. If it detects that you are in a weather state, you can control the wake-up alarm to be automatically canceled (stopped, turned off).
또한, 분석 서버(1200)는, 스마트 워치(1100)로부터 실시간 수신되는 생체 정보를 이용하여 분석된 사용자의 수면단계(현재의 수면단계)를 고려하여, 사용자의 수면 유도를 위해 복수의 뇌파 유도 사운드 중 적어도 하나의 뇌파 유도 사운드를 사용자 단말 또는 스마트 워치(1100)로 제공할 수 있다. 구체적으로, 또한, 분석 서버(1200)는 스마트 워치(1100)로부터 수집된 사용자의 생체 데이터로 분석(파악)된 사용자의 수면단계(현재의 수면단계)에 관한 정보를 기반으로, 최적화된 수면을 유도하기 위해 수면단계별 기 설정된 복수의 뇌파 유도 사운드 중 상기 파악된 사용자의 수면단계에 대응되는 뇌파 유도 사운드가 사용자 단말 또는 스마트 워치(1100)를 통해 재생되도록 제어할 수 있다. 즉, 분석 서버(1200)는 사용자에 대하여 최적화된 수면을 유도하기 위해 사용자의 수면 중 수면단계를 실시간으로 모니터링하여, 수면단계의 변화에 맞추어 수면단계 별로 적합한 뇌파 유도 사운드가 사용자 단말 또는 스마트 워치(1100)에서 자동으로 재생되도록 할 수 있다. 또한, 분석 서버(1200)는 사용자의 생체 데이터를 기반으로 파악된 사용자의 상태(일예로 현재 어느 수면단계에 속하는지에 대한 수면단계 상태)에 따라, 그에 맞는 적합한 뇌파 유도 사운드에 관한 음원 선곡을 자동으로 수행하고, 자동으로 음량 조절을 하여 선곡된 음원이 사용자 단말 또는 스마트 워치(1100)에서 재생되도록 할 수 있다.In addition, the analysis server 1200 considers the user's sleep stage (current sleep stage) analyzed using biometric information received in real time from the smart watch 1100 and generates a plurality of brain wave-induced sounds to induce sleep of the user. At least one brain wave-induced sound may be provided to the user terminal or smart watch 1100. Specifically, the analysis server 1200 provides optimized sleep based on information about the user's sleep stage (current sleep stage) analyzed (identified) with the user's biometric data collected from the smart watch 1100. To induce sleep, among a plurality of preset brain wave induction sounds for each sleep stage, the brain wave induction sound corresponding to the identified user's sleep stage can be controlled to be played through the user terminal or smart watch 1100. That is, the analysis server 1200 monitors the user's sleep stage in real time to induce optimized sleep for the user, and generates brain wave-induced sounds suitable for each sleep stage according to changes in the sleep stage through the user terminal or smart watch ( 1100) can be played automatically. In addition, the analysis server 1200 automatically selects a sound source for an appropriate EEG-induced sound according to the user's status (e.g., which sleep stage the user is currently in) determined based on the user's biometric data. , and automatically adjust the volume so that the selected sound source is played on the user terminal or smart watch 1100.
본 발명에서 뇌파 유도 사운드는 수면을 유도하는 사운드(수면 유도 사운드)로서, 분석 서버(1200)는 사용자의 수면단계별(수면주기의 단계별)로 뇌파 유도 사운드를 재생시킴으로써 사용자의 숙면을 유도할 수 있다.In the present invention, the brain wave-induced sound is a sound that induces sleep (sleep-inducing sound), and the analysis server 1200 can induce a good sleep in the user by playing the brain wave-induced sound according to the user's sleep stage (stage of the sleep cycle). .
여기서, 수면단계별 기 설정된 복수의 뇌파 유도 사운드는, i) 수면단계가 '깨어 있음 또는 거의 깬 상태(즉, 취침 전 단계나 램 수면단계 이후의 기상상태)'일 때 재생되는 제1 뇌파 유도 사운드, ii) 수면단계가 '1단계 수면단계'일 때 재생되는 제2 뇌파 유도 사운드, iii) 수면단계가 '2단계 수면단계'일 때 재생되는 제3 뇌파 유도 사운드, iv) 수면단계가 '3단계 수면단계'일 때 재생되는 제4 뇌파 유도 사운드, v)수면단계가 '4단계 수면단계'일 때 재생되는 제5 뇌파 유도 사운드, 및 vi) 수면단계가 '램(REM) 수면단계'일 때 재생되는 제6 뇌파 유도 사운드를 포함할 수 있다.Here, the plurality of EEG-induced sounds preset for each sleep stage are: i) the first EEG-induced sound played when the sleep stage is 'awake or almost awake (i.e., pre-sleep stage or waking state after REM sleep stage)'; , ii) the 2nd EEG-induced sound played when the sleep stage is 'stage 1 sleep stage', iii) the 3rd EEG-induced sound played when the sleep stage is 'stage 2 sleep stage', iv) the 3rd EEG-induced sound played when the sleep stage is 'stage 3' The 4th EEG-induced sound played when the sleep stage is 'Stage 4 sleep stage', v) The 5th EEG-induced sound played when the sleep stage is 'Stage 4 sleep stage', and vi) The sleep stage is 'REM sleep stage' It may include a sixth brain wave-induced sound that is played when.
이때, 일예로 제1 뇌파 유도 사운드는 낮은 알파파를 유도하는 8Hz 사운드를 의미하고, 제2 뇌파 유도 사운드는 중간 세타파를 유도하는 5Hz 사운드를 의미할 수 있다. 또한, 제4 뇌파 유도 사운드는 델타파를 유도하는 사운드로서, 1Hz 이상 3Hz 이하 중 어느 한 헤르츠(Hz) 값을 갖는 사운드를 의미할 수 있다. 또한, 제6 뇌파 유도 사운드는 낮은 세타파를 유도하는 4Hz 사운드, 및 감마파를 유도하는 사운드(즉, 30Hz 이상 45Hz 이하 중 어느 한 헤르츠 값을 갖는 사운드) 중 적어도 하나를 포함할 수 있다.At this time, as an example, the first brain wave-induced sound may mean an 8Hz sound that induces low alpha waves, and the second brain wave-induced sound may mean a 5Hz sound that induces medium theta waves. In addition, the fourth brain wave-induced sound is a sound that induces delta waves, and may mean a sound having any one hertz (Hz) value between 1Hz and 3Hz. Additionally, the sixth brain wave-induced sound may include at least one of a 4Hz sound that induces low theta waves and a sound that induces gamma waves (i.e., a sound having any one hertz value between 30Hz and 45Hz).
다시 말하면, 예시적으로 분석 서버(1200)는, 사용자가 취침 전 또는 기상상태인 것으로 파악될 때 낮은 알파파를 유도하는 8Hz 사운드(즉, 제1 뇌파 유도 사운드)가 스마트폰 또는 스마트 워치(1100)에서 자동으로 재생되도록 할 수 있다. 여기서, 낮은 알파파는 취침 전과 기상 직전의 자연적인 뇌파 주파수를 모방하며, 이에 따라 분석 서버(1200)는 낮은 알파파를 유도하는 제1 뇌파 유도 사운드를 제공함으로써, 사용자가 부드럽게 수면에 들고 수면에서 깨어날 수 있도록 도와줄 수 있다.In other words, by way of example, the analysis server 1200 provides an 8Hz sound (i.e., a first brain wave-induced sound) that induces low alpha waves when the user is determined to be in a state before going to bed or waking up. ) can be played automatically. Here, low alpha waves mimic natural brain wave frequencies before going to bed and just before waking up, and accordingly, the analysis server 1200 provides a first brain wave inducing sound that induces low alpha waves, allowing the user to gently fall asleep and wake up from sleep. can help you do it.
또한, 분석 서버(1200)는, 사용자가 1단계 수면단계인 것으로 파악될 때 중간 세타파를 유도하는 5Hz 사운드(즉, 제2 뇌파 유도 사운드)가 스마트폰 또는 스마트 워치(1100)에서 자동으로 재생되도록 할 수 있다. 여기서, 중간 세타파는 긴장을 완화시켜 사용자가 편안하게 수면을 취할 수 있도록 도와줄 수 있다.In addition, the analysis server 1200 automatically plays a 5Hz sound (i.e., a second brain wave-induced sound) that induces intermediate theta waves on the smartphone or smart watch 1100 when it is determined that the user is in stage 1 sleep. can do. Here, the middle theta waves can help users sleep comfortably by relieving tension.
또한, 분석 서버(1200)는, 사용자가 3단계 수면단계인 것으로 파악될 때 델타파를 유도하는 1~3Hz 사운드(즉, 제4 뇌파 유도 사운드)가 스마트폰 또는 스마트 워치(1100)에서 자동으로 재생되도록 할 수 있다. 델타파는 수면 주기에서 가장 깊이 잠들 수 있도록 도와줄 수 있다.In addition, when the analysis server 1200 determines that the user is in stage 3 sleep, a 1 to 3 Hz sound (i.e., fourth brain wave inducing sound) that induces delta waves is automatically generated from the smartphone or smart watch 1100. You can make it play. Delta waves can help you fall into the deepest sleep of your sleep cycle.
또한, 분석 서버(1200)는, 사용자가 램 수면단계인 것으로 파악될 때 낮은 세타파를 유도하는 4Hz 사운드와 감마파를 유도하는 30~45Hz 사운드가 스마트폰 또는 스마트 워치(1100)에서 자동으로 재생되도록 할 수 있다. 상술한 본 발명의 일예에서는 분석 서버(1200)가 사용자의 수면 유도를 위해 뇌파 유도 사운드(즉, 소리, 음원, 사운드)를 제공하는 것으로 예시하였으나, 이에만 한정되는 것은 아니고, 수면 유도를 위해 수면 유도 영상 등의 다양한 형태의 콘텐츠를 제공할 수도 있다.In addition, the analysis server 1200 automatically plays the 4Hz sound that induces low theta waves and the 30-45Hz sound that induces gamma waves on the smartphone or smart watch (1100) when it is determined that the user is in the REM sleep stage. can do. In an example of the present invention described above, the analysis server 1200 is illustrated as providing brain wave-induced sound (i.e., sound, sound source, sound) to induce sleep in the user, but the present invention is not limited thereto, and the analysis server 1200 provides sleep inducing sound to the user. Various types of content, such as guided videos, can also be provided.
이에 따르면, 본 발명의 분석 서버(1200)는 스마트 워치(1100)를 이용하여 실시간으로 사용자의 수면단계를 모니터링함으로써 라이트 테라피, 뇌파 유도 등의 방법을 통해 실시간으로 수면을 코칭할 수 있다.According to this, the analysis server 1200 of the present invention can provide sleep coaching in real time through methods such as light therapy and brain wave induction by monitoring the user's sleep stage in real time using the smart watch 1100.
또한, 분석 서버(1200)는 산출된 기상시간 정보를 고려하여 스마트 워치(1100)에 대한 블루라이트 차단 모드의 ON/OFF와 블루라이트 증강 모드의 ON/OFF를 제어할 수 있다. 이때, 본 발명에서 고려되는 블루라이트 증강 모드는 블루라이트 차단 모드의 ON/OFF 기능과는 다르게, 아침에 사용자가 사용자 단말(일예로 스마트폰 등)을 사용할 때 사용자 단말의 디스플레이 화면을 전반적으로 푸르게 하는 기능을 의미할 수 있다. 이에 대한 보다 구체적인 설명은 다음과 같다.Additionally, the analysis server 1200 can control ON/OFF of the blue light blocking mode and ON/OFF of the blue light enhancement mode for the smart watch 1100 in consideration of the calculated wake-up time information. At this time, the blue light enhancement mode considered in the present invention, unlike the ON/OFF function of the blue light blocking mode, makes the display screen of the user terminal blue overall when the user uses the user terminal (for example, a smartphone, etc.) in the morning. It can mean a function that does. A more specific explanation for this is as follows.
한편, 도 14는 스마트폰의 빛 스펙트럼을 설명하기 위한 도면이다. 도 14를 참조하면, 일예로 멜라토닌(melatonin)은 수면호르몬으로서, 낮에는 적게 분비되고, 밤에는 많이 분비되면서 수면을 조절 및 유도하여 하루 주기 리듬에 작용한다. 건강한 하루 주기 리듬에서 멜라토닌은 낮 동안에 억제되었다가 밤에 생성되어 숙면을 취하게 한다. 우리몸은 오전 시간에 햇빛의 파장인 400∼ nm 스펙트럼의 빛을 인식하여 멜라토닌을 억제한다. 또한, 블루라이트는 스마트폰 등에서 나오는 380∼ nm의 파장을 가진 파란색 계열의 가시광선을 의미하는데, 이러한 블루라이트는 밤에도 낮인 것처럼 착각하게 해 사용자의 생체 시계 리듬을 흐트려 놓는 특성이 있다.Meanwhile, Figure 14 is a diagram for explaining the light spectrum of a smartphone. Referring to FIG. 14, for example, melatonin is a sleep hormone that is secreted in small amounts during the day and in large quantities at night, thereby controlling and inducing sleep and acting on the circadian rhythm. In a healthy circadian rhythm, melatonin is suppressed during the day and produced at night to promote sound sleep. In the morning, our body recognizes light in the 400 nm spectrum, which is the wavelength of sunlight, and suppresses melatonin. In addition, blue light refers to blue visible light with a wavelength of 380 nm emitted from smartphones, etc. This blue light has the characteristic of disrupting the user's biological clock rhythm by making the user think it is daytime even at night.
이에, 분석 서버(1200)는 앞서 말한 바와 같이 수면효율을 최적화해주는 최적 취침시간과 최적 기상시간을 산출하여, 산출된 최적 취침시간을 사용자의 취침알람 시간으로 설정하고, 산출된 최적 기상시간을 사용자의 기상알람 시간으로 설정할 수 있다. 분석 서버(1200)는 최적 취침시간에 대응하는 시간(즉, 취침알람 시간)에 취침알람이 스마트 워치(1100)를 통해 제공되도록 하고, 최적 기상시간에 대응하는 시간(즉, 기상알람 시간)에 기상알람이 스마트 워치(1100)를 통해 제공되도록 할 수 있다. 이때, 예시적으로 최적 취침시간이 오후 11시이고, 최적 기상시간이 오전 4시로 산출되었다고 하자.Accordingly, as mentioned above, the analysis server 1200 calculates the optimal bedtime and optimal wake-up time that optimize sleep efficiency, sets the calculated optimal bedtime as the user's bedtime alarm time, and sets the calculated optimal wake-up time to the user's bedtime. You can set the wake-up alarm time. The analysis server 1200 ensures that a sleep alarm is provided through the smart watch 1100 at a time corresponding to the optimal bedtime (i.e., bedtime alarm time), and at a time corresponding to the optimal wake-up time (i.e., wake-up alarm time). A wake-up alarm can be provided through the smart watch 1100. At this time, let us assume that the optimal bedtime is 11 PM and the optimal wake-up time is calculated to be 4 AM.
이때, 분석 서버(1200)는 최적 취침시간 이전(일예로 최적 취침시간으로부터 기 설정된 시간 이전)부터 최적 기상시간 이전에 해당하는 밤 시간 구간 동안에(일예로 오후 8시부터 오전 4시까지) 스마트 워치(1100) 및 사용자 단말(미도시) 각각이 블루라이트 차단 모드로 작동하도록 제어(즉, 블루라이트 차단 모드를 ON으로 제어)할 수 있다. At this time, the analysis server 1200 uses a smart watch during the night time section corresponding to before the optimal bedtime (for example, before a preset time from the optimal bedtime) to before the optimal wake-up time (for example, from 8 PM to 4 AM). (1100) and the user terminal (not shown) can each be controlled to operate in the blue light blocking mode (i.e., control the blue light blocking mode to ON).
반면, 분석 서버(1200)는 사용자가 스마트 워치(1100) 및 사용자 단말(미도시)을 아침에 사용하는 경우, 스마트 워치(1100) 및 사용자 단말(미도시) 각각에 대하여 블루라이트 증강 모드를 ON으로 제어할 수 있다. 여기서, 아침은 예시적으로 상술한 밤 시간 구간을 제외한 시간 구간(일예로 오전 4시부터 오후 8시 사이에 해당하는 낮 시간 구간) 중 일부 시간 구간을 의미할 수 있다. 일예로, 분석 서버(1200)는 스마트 워치(1100) 및 사용자 단말(미도시)에 대하여, 밤 시간 구간일 때 블루라이트 차단 모드를 ON 시킴과 더불어 블루라이트 증강 모드를 OFF 시킬 수 있고, 낮 시간 구간일 때 블루라이트 차단 모드를 OFF 시킴과 더불어 블루라이트 증강 모드를 ON 시킬 수도 있다.On the other hand, when the user uses the smart watch 1100 and the user terminal (not shown) in the morning, the analysis server 1200 turns on the blue light enhancement mode for each of the smart watch 1100 and the user terminal (not shown). It can be controlled with . Here, morning may illustratively mean some of the time sections (for example, the daytime section between 4:00 AM and 8:00 PM) excluding the night time section described above. As an example, the analysis server 1200 can turn on the blue light blocking mode and turn off the blue light enhancement mode for the smart watch 1100 and the user terminal (not shown) during the night time section, and turn off the blue light enhancement mode during the day. In addition to turning off the blue light blocking mode, you can also turn on the blue light enhancement mode.
분석 서버(1200)는 이러한 블루라이트 증강 모드 기능을 제공함으로써, 아침에(특히, 상술한 낮 시간 구간 동안에) 사용자가 사용자 단말(스마트폰, 휴대폰)이나 스마트 워치를 사용할 때 디스플레이에서 파란색 계열의 색이 포함된 빛을 발광하는 모드(즉, 후술하여 설명하는 제1 파장 범위에 속하는 파장의 빛이 발광하도록 하는 모드)로 제어되도록 해, 아침(오전)에 사용자를 각성시키는 효과를 제공하여 사용자가 원하는 하루 주기 리듬을 설정할 수 있도록 할 수 있다. 이는 광선치료(광치료)에서 블루라이트의 경우 빛의 세기와 관계없이 30분 이상 1시간 이하의 시간 정도만 쐬어도 충분하기 때문이다.The analysis server 1200 provides this blue light enhancement mode function, so that when a user uses a user terminal (smart phone, mobile phone) or smart watch in the morning (particularly during the above-mentioned daytime period), a blue-based color is displayed on the display. The included light is controlled in a mode that emits light (i.e., a mode that emits light with a wavelength belonging to the first wavelength range, which will be described later), providing the effect of awakening the user in the morning (morning), thereby allowing the user to You can set your desired circadian rhythm. This is because in the case of phototherapy (light therapy), exposure to blue light for at least 30 minutes and less than 1 hour is sufficient, regardless of the intensity of the light.
여기서, 블루라이트 증강 모드는, 사용자에게 각성 효과의 제공 및 멜라토닌 분비 억제를 위해, 스마트 워치(1100)와 사용자 단말 각각의 화면(LCD 화면) 및 LED 표시부에서, 380 nm 이상 500 nm 이하에 해당하는 제1 파장 범위 중 어느 한 파장의 빛이 제공(발광)되도록 하는 모드(즉, 파란색 계열의 색이 포함된 빛을 발광하는 모드)를 의미할 수 있다. 반면, 블루라이트 차단 모드는, 사용자에게 숙면 효과의 제공 및 멜라토닌 생성(분비)을 위해, 스마트 워치(1100)와 사용자 단말 각각의 화면(LCD 화면) 및 LED 표시부에서, 500 nm 초과 700 nm 이하에 해당하는 제2 파장 범위 중 어느 한 파장의 빛이 제공(발광)되도록 하는 모드(즉, 주황색 계열의 색이 포함된 빛을 발광하는 모드)를 의미할 수 있다.Here, the blue light enhancement mode is to provide an awakening effect to the user and suppress melatonin secretion, corresponding to 380 nm to 500 nm on the screen (LCD screen) and LED display unit of the smart watch 1100 and the user terminal, respectively. This may mean a mode in which light of any one wavelength in the first wavelength range is provided (emitted) (i.e., a mode in which light containing a blue color is emitted). On the other hand, in the blue light blocking mode, in order to provide the user with a good sleep effect and produce (secrete) melatonin, the screen (LCD screen) and LED display unit of the smart watch (1100) and the user terminal respectively exceed 500 nm and below 700 nm. It may refer to a mode in which light of any one wavelength in the corresponding second wavelength range is provided (emitted) (i.e., a mode in which light containing an orange-based color is emitted).
상술한 설명에서는, 일예로 분석 서버(1200)가 사용자가 소유한 스마트 워치(1100) 및 사용자 단말을 대상으로 블루라이트 차단 모드 또는 블루라이트 증강 모드의 ON/OFF를 제어하는 것으로 예시하였으나, 이에만 한정되는 것은 아니고, 분석 서버(1200)는 사용자가 소유한 각종 전자기기(일예로, VR 기기, 컴퓨터 모니터, TV 등)의 디스플레이(화면)를 대상으로 블루라이트 차단 모드의 ON/OFF를 제어할 수도 있다. 이를 위해, 일예로 사용자는 사용자 단말을 통해 분석 서버(1200)에 블루라이트 차단 모드 또는 블루라이트 증강 모드의 ON/OFF 제어를 희망하는 전자기기들을 미리 등록시켜 둘 수 있다.In the above description, as an example, the analysis server 1200 controls the ON/OFF of the blue light blocking mode or blue light enhancement mode for the smart watch 1100 and the user terminal owned by the user, but only in this case. It is not limited, and the analysis server 1200 can control the ON/OFF of the blue light blocking mode for the display (screen) of various electronic devices (for example, VR devices, computer monitors, TVs, etc.) owned by the user. It may be possible. To this end, for example, the user may pre-register electronic devices for which they wish to control the ON/OFF of the blue light blocking mode or blue light enhancement mode in the analysis server 1200 through the user terminal.
이러한, 분석 서버(1200)는 블루라이트 차단 모드 또는 블루라이트 증강 모드의 ON/OFF 제어 기능을 제공함으로써, 낮 시간 구간 동안에는 제1 파장 범위에 속하는 스펙트럼 파장의 빛을 블루라이트로 발광시켜 각성 효과를 제공하고, 밤 시간 구간 동안에는 제2 파장 범위에 속하는 스펙트럼 파장의 빛을 발광시켜 숙면 효과를 제공하는, 광선치료(light therapy) 기능을 구현할 수 있다.The analysis server 1200 provides an ON/OFF control function of the blue light blocking mode or blue light enhancement mode, thereby providing an awakening effect by emitting light with a spectrum wavelength belonging to the first wavelength range as blue light during the daytime section. In addition, a light therapy function that provides a sound sleep effect by emitting light of a spectrum wavelength belonging to the second wavelength range during the night time period can be implemented.
다시 말해, 분석 서버(1200)는 블루라이트 차단 모드를 제공할 수 있다. 밤에도 낮인 것처럼 착각하게 해 생체 시계 리듬을 흐트려 놓는 블루라이트를 프로그램으로 차단하는 원리는 디지털 화면을 구성하는 RGB (빨강, 초록, 파랑) 색상 중 파란색 계열의 광원을 줄이는 방식이다.In other words, the analysis server 1200 may provide a blue light blocking mode. The principle of blocking blue light, which disrupts the biological clock rhythm by making people think it is daytime even at night, with a program is to reduce the blue light source among the RGB (red, green, blue) colors that make up the digital screen.
기존의 블루라이트 차단 모드는 야간모드(다크모드)를 적용한 것처럼 디스플레이(화면)가 어두워지면서 따뜻한 색으로 변하게 된다. 또한, 기존에는 프로그램에 따라 사용자가 원하는 정도(즉, 블루라이트 차단 강도)와 원하는 시간(직접 지정한 특정 시간)에 블루라이트가 차단되도록 설정(세팅)하는 것이 가능하기도 하다.In the existing blue light blocking mode, the display (screen) darkens and changes to warm colors, as if applying night mode (dark mode). In addition, depending on the existing program, it is possible to set the blue light to be blocked to the extent desired by the user (i.e., blue light blocking intensity) and at the desired time (a specific time designated directly).
그리고, 기존의 야간모드(다크모드) 기능은 근시와 난시를 유발하는 특성이 있다. 즉, 기존의 야간모드(다크모드)는, 어두운 배경에 밝은 글씨가 나타나는 사용자환경(UI)을 제공하는데, 이와 같은 사용자환경(UI)에서 빛이 줄어들면 동공이 확장되고 그 안에 들어온 빛들이 한곳에 모이지 못해 깨끗하고 선명한 상을 만들지 못하게 됨에 따라, 눈 안의 수정체의 위치가 앞으로 이동하게 되어 근시를 유발하게 되는 경향이 있다. Additionally, the existing night mode (dark mode) function has the characteristic of causing myopia and astigmatism. In other words, the existing night mode (dark mode) provides a user environment (UI) where bright text appears on a dark background. In this user environment (UI), when the light decreases, the pupil expands and the light entering the pupil is concentrated in one place. As the lens cannot come together to create a clean and clear image, the position of the lens inside the eye tends to move forward, causing myopia.
이에, 본 발명에서는 상술한 기존의 블루라이트 차단 모드의 설정 방식 및 종래의 야간모드(다크모드)가 갖는 단점을 개선하여, 알고리즘을 통해 사용자 개인별 취침시간에 적합한 시간이 되면 자동으로 블루라이트 차단이 이루어지도록 설정(제어)할 수 있다. 즉, 분석 서버(1200)는 사용자의 생체 데이터의 분석을 토대로 산출된 최적 취침시간과 최적 기상시간을 기반으로, 최적 취침시간에 해당하는 시간이 되면 스마트 워치(1100) 및 사용자 단말에 대해 자동으로 블루라이트 차단 모드를 ON으로 제어하고, 이후 시간이 지나 최적 기상시간에 해당하는 시간이 되면(도래하면) 스마트 워치(1100) 및 사용자 단말에 대해 자동으로 블루라이트 차단 모드를 OFF 시킴으로써 블루라이트 모드가 ON되도록 제어할 수 있다.Accordingly, the present invention improves the disadvantages of the existing blue light blocking mode setting method and the conventional night mode (dark mode) described above, and automatically blocks blue light when the time appropriate for the user's individual bedtime is reached through an algorithm. You can set (control) it to happen. That is, the analysis server 1200 automatically updates the smart watch 1100 and the user terminal when the time corresponding to the optimal bedtime arrives, based on the optimal bedtime and optimal wake-up time calculated based on analysis of the user's biometric data. The blue light blocking mode is controlled to ON, and when the time corresponding to the optimal wake-up time arrives (arrival), the blue light blocking mode is automatically turned OFF for the smart watch (1100) and the user terminal, thereby turning the blue light mode on. It can be controlled to be ON.
종래에는 블루라이트의 ON/OFF 제어에 대한 시간 설정이 사용자에 의해 직접 지정된(즉, 수동으로 직접 지정된) 시간으로 설정되었던 반면, 본 발명에서 분석 서버(1200)는 자동으로 산출된 시간(즉, 최적 최침시간과 최적 기상시간)을 토대로 자동 설정될 수 있다.Conventionally, the time setting for ON/OFF control of blue light was set to a time directly specified by the user (i.e., manually specified), whereas in the present invention, the analysis server 1200 automatically calculates the time (i.e. It can be set automatically based on optimal sleep time and optimal wake-up time.
또한, 분석 서버(1200)는, 사용자가 기상하는 시간이 도래했을 때(즉 최적 기상시간의 직전에), 스마트 워치(1100) 및 사용자 단말의 화면(디스플레이)에서 자동으로 일예로 480nm의 스펙트럼 파장의 푸른 빛이 발광(제공)되도록 하여 각성 효과를 제공할 수 있다. 이후, 분석 서버(1200)는 아침 시간 동안에 블루라이트 증강 모드를 ON으로 설정할 수 있다. 이후, 분석 서버(1200)는 사용자가 취침하는 시간이 도래했을 때(즉, 최적 취침시간의 직전에), 스마트 워치(1100) 및 사용자 단말의 화면(디스플레이)에서 자동으로 일예로 580 nm의 스펙트럼 파장의 주황 빛이 발광(제공)되도록 하여 숙면 효과를 제공할 수 있다. 또한, 분석 서버(1200)는 후술하여 설명하는 것과 같이 사용자 단말의 후면에 있는 Flash light LED에 블루라이트 차단 필름을 부착한 후 밤에 휴대폰 사용시 후면 Flash light LED를 발광시키도록 제어할 수 있다. In addition, the analysis server 1200 automatically displays a spectrum wavelength of 480 nm on the screen (display) of the smart watch 1100 and the user terminal when the time for the user to wake up arrives (i.e., just before the optimal wake-up time). It can provide an awakening effect by emitting (providing) blue light. Thereafter, the analysis server 1200 may set the blue light enhancement mode to ON during morning hours. Thereafter, when the user's bedtime arrives (i.e., just before the optimal bedtime), the analysis server 1200 automatically displays, for example, a spectrum of 580 nm on the screen (display) of the smart watch 1100 and the user terminal. It can provide a sound sleep effect by emitting (providing) orange light of a certain wavelength. In addition, the analysis server 1200 can attach a blue light blocking film to the flash light LED on the rear of the user terminal and control the rear flash light LED to emit light when using the mobile phone at night, as will be described later.
블루라이트 증강 모드 기능의 제공에 의하면, 분석 서버(1200)는 아침에(특히, 상술한 낮 시간 구간 동안에) 사용자가 사용자 단말(스마트폰, 휴대폰)을 사용할 때 디스플레이의 블루라이트를 프로그램으로 증강시킴으로써(이때, 블루라이트를 증강시켰다 함은 블루라이트 차단 모드를 OFF로 제어한 상태를 의미할 수 있음), 아침(오전)에 사용자를 각성시키는 효과를 제공하여 사용자가 원하는 하루 주기 리듬을 설정할 수 있도록 할 수 있다. 이는 광선치료(광치료)에서 블루라이트의 경우 빛의 세기와 관계없이 30분 이상 1시간 이하의 시간 정도만 쐬어도 충분하기 때문이다. 반면, 백색광은 빛의 세기에 따라 1만 룩스(lux)는 30분, 2500 룩스(lux)는 2시간 정도 쐬는 것이 권장된다.According to the provision of the blue light enhancement mode function, the analysis server 1200 enhances the blue light of the display with a program when the user uses the user terminal (smart phone, mobile phone) in the morning (especially during the above-mentioned daytime section). (At this time, enhanced blue light may mean that the blue light blocking mode is controlled to OFF), providing the effect of awakening the user in the morning (morning) so that the user can set the desired circadian rhythm. can do. This is because in the case of phototherapy (light therapy), exposure to blue light for at least 30 minutes and less than 1 hour is sufficient, regardless of the intensity of the light. On the other hand, depending on the intensity of white light, it is recommended to use 10,000 lux for 30 minutes and 2,500 lux for 2 hours.
또한, 분석 서버(1200)는 블루라이트 차단을 위한 블루라이트 차단 플래시 라이트 LED(Flash light LED)를 제공할 수 있다. 구체적으로, 사용자의 눈을 보호하면서 밤에 핸드폰을 사용하는 가장 좋은 방법은 밤에 불을 끈 상태에서 휴대폰을 보지 않는 것이라 할 수 있다. 즉, 조명을 켠 밝은 곳에서 핸드폰의 원래 화면으로 보는 것이 사람의 눈을 망가지지 않게 하는 가장 좋은 방법이라 할 수 있다. 하지만, 이처럼 조명이 켜져 있는 밝은 곳에서의 핸드폰 사용은 핸드폰과 조명의 블루라이트로 인해 사용자의 수면을 방해하게 되는 문제가 있다.Additionally, the analysis server 1200 may provide a blue light blocking flash light LED for blocking blue light. Specifically, the best way to use a cell phone at night while protecting the user's eyes is to not look at the cell phone at night with the lights off. In other words, viewing the original screen of a cell phone in a bright place with the lights turned on is the best way to avoid damaging one's eyes. However, using a cell phone in a bright place with the lights on like this has the problem of disrupting the user's sleep due to the blue light from the cell phone and lights.
이러한 문제를 개선하기 위해, 일예로, 분석 서버(1200)는 사용자가 소지한 사용자 단말의 후면에 있는 플래시 라이트 LED 상에 블루라이트 차단 필름이 부착되도록 하고, 이후 밤에(특히, 상술한 밤 시간 구간 동안에) 사용자가 사용자 단말을 사용하는 것으로 감지되면, 사용자 단말의 후면의 플래시 라이트 LED가 발광되도록 제어하여, 이를 통해 블루라이트가 차단된 빛이 제공되도록 할 수 있다. 다른 예로, 분석 서버(1200)는 사용자 단말의 후면에 분석 서버(1200)에 의해 제공되는 자체 개발된 블루라이트 차단 플래시 라이트 LED가 설치되도록 할 수 있다.In order to improve this problem, for example, the analysis server 1200 attaches a blue light blocking film to the flash light LED on the back of the user terminal carried by the user, and then at night (particularly during the night time described above). (during the section), when it is detected that the user is using the user terminal, the flash light LED on the back of the user terminal is controlled to emit light, thereby providing light with blue light blocked. As another example, the analysis server 1200 may install a self-developed blue light blocking flash light LED provided by the analysis server 1200 on the rear of the user terminal.
이때, 자체 개발된 블루라이트 차단 플래시 라이트 LED는, 일예로 블루라이트 차단 필름이 기본 플래시 라이트 LED 상에 부착된 형태로 마련되되, 사용자 단말에 설치 가능한 형태로 마련되는 것일 수 있다. 또한, 블루라이트 차단 플래시 라이트 LED는 일예로 제2 파장 범위(즉, 500 nm 초과 700 nm 이하의 파장 범위)에 속하는 파장의 빛을 발광(조사, 제공)하도록 마련되는 것일 수 있다. 이러한 분석 서버(1200)는 블루라이트 차단 플래시 라이트 LED를 제공함으로써, 사용자가 밤에 휴대폰 사용이 불가피한 경우에도 눈을 보호하면서 휴대폰 사용이 가능하도록 할 수 있다.At this time, the self-developed blue light blocking flash light LED may, for example, be provided in a form in which a blue light blocking film is attached to the basic flash light LED, and may be installed in a user terminal. In addition, the blue light blocking flash light LED may, for example, be prepared to emit (irradiate, provide) light of a wavelength belonging to a second wavelength range (i.e., a wavelength range of more than 500 nm and less than 700 nm). This analysis server 1200 provides a blue light blocking flash light LED, allowing the user to use the mobile phone while protecting the eyes even when use of the mobile phone is unavoidable at night.
즉, 분석 서버(1200)는 블루라이트 차단 모드의 ON/OFF 제어 기능, 블루라이트 증강 모드 기능(즉, 블루라이트 증강 ON/OFF 기능), 및 블루라이트 차단 플래시 라이트 LED를 제공함으로써, 사용자 단말(즉, 휴대폰)이나 스마트 워치(1100) 등의 사용으로 인한 사용자의 근시 및 난시의 유발을 효과적으로 예방되도록 할 수 있다.That is, the analysis server 1200 provides a blue light blocking mode ON/OFF control function, a blue light enhancement mode function (i.e., a blue light enhancement ON/OFF function), and a blue light blocking flash light LED, thereby providing a user terminal ( In other words, it is possible to effectively prevent the user from experiencing myopia and astigmatism due to use of a mobile phone or smart watch 1100.
분석 서버(1200)는 야간모드(다크모드)의 제공 대신, 블루라이트 차단 기능(특히, 디스플레이 화면을 제어하거나 사용자 단말의 후면에 블루라이트 차단 플래시 라이트 LED를 설치함에 따른 블루라이트 차단 기능)을 제공할 수 있다.Instead of providing a night mode (dark mode), the analysis server 1200 provides a blue light blocking function (in particular, a blue light blocking function by controlling the display screen or installing a blue light blocking flash light LED on the back of the user terminal). can do.
또한, 분석 서버(1200)는 본 발명에서 고려되는 다양한 디스플레이 디바이스(일예로, 스마트폰, TV, 모니터, 스마트 워치 등)의 후면과 측면에 블루라이트를 차단한 500~700nm 파장(즉, 제2 파장 범위)의 빛을 발광하는 배경 LED(달리 표현해 배경 라이트 LED)를 설치함으로써, 사용자가 디스플레이 디바이스를 야간에 사용할 시에 수면을 방해하지 않으면서도 근시와 난시 발생을 예방하면서 사용 가능하도록 할 수 있다. 즉, 분석 서버(1200)는 사용자가 야간에 디스플레이 디바이스 사용시 발생할 수 있는 근시와 난시를 예방하기 위해 디스플레이 디바이스의 후면과 측면에 하드웨어적으로 또는 탈부착 식의 LED(이는 상술한 배경 LED나 블루라이트 차단 플래시 라이트 LED를 의미함)를 설치 가능하도록 할 수 있다. 여기서, 디스플레이 디바이스는, 본 발명에서 고려되는 각종 단말 기기를 의미하는 것으로서, 사용자 단말(일예로 스마트폰 등), 스마트 워치, TV, 모니터 등의 기기를 포함할 수 있다.In addition, the analysis server 1200 is provided with a 500-700nm wavelength (i.e., a second wavelength that blocks blue light) on the back and sides of various display devices (e.g., smartphones, TVs, monitors, smart watches, etc.) considered in the present invention. By installing a background LED (in other words, background light LED) that emits light in the wavelength range), it is possible to prevent the occurrence of myopia and astigmatism without disturbing sleep when the user uses the display device at night. . That is, the analysis server 1200 installs hardware or detachable LEDs on the back and sides of the display device to prevent myopia and astigmatism that may occur when the user uses the display device at night (this blocks the background LED or blue light described above). Flash light (meaning LED) can be installed. Here, the display device refers to various terminal devices considered in the present invention and may include devices such as user terminals (eg, smartphones, etc.), smart watches, TVs, and monitors.
다시 말해, 분석 서버(1200)는 스마트 워치 및 사용자가 소지한 사용자 단말을 포함한 디스플레이 디바이스의 일영역(일예로 후면과 측면)에 설치 가능한 배경 라이트 LED를 사용자에게 제공할 수 있고, 여기서, 배경 라이트 LED는 사용자가 야간에 디스플레이 디바이스를 사용해야 할 경우, 디스플레이 디바이스의 사용으로 인한 근시와 난시 발생의 예방이 가능한 형태로 마련될 수 있다. 배경 라이트 LED는 디스플레이 디바이스에 대하여 하드웨어적으로 설치되거나 탈부착 방식으로 설치 가능한 형태로 마련될 수 있다. 본 발명의 분석 서버(1200)는 야간에 디스플레이 디바이스 사용시 발생할 수 있는 근시, 난시, 수면장애를 디스플레이 애플리케이션(Application) 소프트웨어 및 하드웨어를 통해 예방할 수 있다.In other words, the analysis server 1200 can provide the user with a background light LED that can be installed in one area (for example, the back and sides) of the display device, including the smart watch and the user terminal carried by the user, where the background light LEDs can be provided in a form that can prevent myopia and astigmatism due to use of the display device when a user must use the display device at night. The background light LED can be installed as hardware to the display device or can be installed in a detachable manner. The analysis server 1200 of the present invention can prevent myopia, astigmatism, and sleep disorders that may occur when using a display device at night through display application software and hardware.
즉, 본 발명에서 고려되는 배경 라이트 LED(즉, 배경 LED, 블루라이트 차단 플래시 라이트 LED)는 근시와 난시 예방이 가능한 형태로 마련되고, 스마트 워치와 사용자 단말을 포함하여 사용자에 의해 기 등록된 디스플레이 디바이스에 설치 가능하게 마련될 수 있다. 또한, 상기 디스플레이 디바이스에 설치되는 배경 라이트 LED(즉, 배경 LED, 블루라이트 차단 플래시 라이트 LED)와 상기 디스플레이 디바이스의 후면에 마련된 플래시 라이트 LED(즉, 통상적인 라이트인 기본 플래시 라이트 LED)는, 상기 분석 서버(1200)에 의한 제어 또는 사용자 입력에 기반한 제어에 의해 사용, 설정 및 작동이 가능하도록 마련될 수 있다. 즉, 본 발명에서, 근시와 난시 예방을 위해 마련되는 배경 라이트 LED와, 디스플레이 디바이스의 후면에 있는 플레시 라이트 LED(통상적인 라이트)는, 이들을 가동하게 하는 애플리케이션 등의 소프트웨어의 인간에 의한 설정에 의해 제어되도록 마련될 수 있고, 또한 이들은 본 발명에서 제안하는 실시간 수면 코칭 기술에서 뿐만 아니라 그밖에 일반적으로도 사용 가능하도록 마련될 수 있다.That is, the background light LED (i.e., background LED, blue light blocking flash light LED) considered in the present invention is provided in a form capable of preventing myopia and astigmatism, and is used on displays pre-registered by the user, including smart watches and user terminals. It may be provided to be installed on a device. In addition, the background light LED installed in the display device (i.e., background LED, blue light blocking flash light LED) and the flash light LED provided on the rear of the display device (i.e., basic flash light LED, which is a typical light) are, It may be arranged to enable use, setting, and operation by control by the analysis server 1200 or control based on user input. That is, in the present invention, the background light LED provided to prevent myopia and astigmatism and the flash light LED (ordinary light) on the back of the display device are controlled by human settings of software such as an application that operates them. They can be arranged to be controlled, and they can also be arranged to be usable not only in the real-time sleep coaching technology proposed by the present invention but also in other general applications.
또한, 배경 라이트 LED는 사용자에 의해 기 등록된 디스플레이 디바이스 뿐만 아니라 디스플레이 디바이스의 제조시에도 근시와 난시 예방이 가능한 배경 라이트 LED의 설치가 가능하도록 마련될 수 있다.In addition, the background light LED may be provided so that the background light LED capable of preventing myopia and astigmatism can be installed not only in the display device pre-registered by the user but also when manufacturing the display device.
여기서, 사용자에 의해 기 등록된 디스플레이 디바이스에는 스마트 워치, 사용자 단말(스마트폰 등)을 포함하여 사용자가 소유한 각종 전자기기(일예로, VR 기기, 컴퓨터 모니터, TV 등의 다양한 종류의 디스플레이 디바이스)가 포함될 수 있다. 이에 따르면, 본 발명은 본 시스템(11)(즉, AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템) 외에 일반적인 상황에서 사용자가 디스플레이 디바이스를 사용할 경우에도 근시와 난시 예방이 가능한 배경 라이트 LED의 설치가 가능하도록 할 수 있는바, 디스플레이 디바이스의 사용으로 인한 근시와 난시의 발생이 효과적으로 예방(방지)되도록 할 수 있다.Here, the display device pre-registered by the user includes various electronic devices owned by the user, including smart watches and user terminals (smartphones, etc.) (for example, various types of display devices such as VR devices, computer monitors, TVs, etc.) may be included. According to this, in addition to the present system 11 (i.e., a real-time sleep health management service providing system using AI-based brain wave tuning and autonomic nervous system control), the present invention is capable of preventing myopia and astigmatism even when a user uses a display device in a normal situation. Since it is possible to install a background light LED, the occurrence of myopia and astigmatism due to the use of the display device can be effectively prevented (prevented).
기존의 블루라이트 차단 모드는 ON으로 제어되었을 때 디스플레이가 어두워지면서 400~500nm의 블루라이트를 차단하여 색상 스펙트럼의 따뜻한 색으로 변환시키게 되는데, 이 또한 디스플레이가 어두워지기 때문에 근시와 난시를 유발하는 단점이 있다. 이를 개선하기 위해 분석 서버(1200)는 디스플레이 디바이스를 밝게 한 상태에서(즉, 디스플레이 화면이 기 설정된 밝기값 이상의 밝기를 갖는 상태에서) 블루라이트 차단 모드를 작동(ON)시킬 수 있다.The existing blue light blocking mode darkens the display when turned ON, blocking blue light of 400 to 500 nm and converting it to a warmer color in the color spectrum. However, this also darkens the display, which has the disadvantage of causing myopia and astigmatism. there is. To improve this, the analysis server 1200 may turn on the blue light blocking mode while the display device is brightened (that is, the display screen has a brightness higher than a preset brightness value).
또한 분석 서버(1200)는 일예로 디스플레이 디바이스의 조도센서(광센서)를 이용하여 디스플레이 디바이스의 주변 빛의 양을 측정하고, 측정된 주변 빛의 양을 기반으로 상술한 배경 LED의 발광의 작동을 제어할 수도 있다. 즉, 분석 서버(1200)는 디스플레이 디바이스의 조도센서(광센서)를 활용하여 주변 빛의 양을 측정하여 배경 LED 발광을 연동할 수 있다.In addition, the analysis server 1200, for example, measures the amount of ambient light of the display device using an illumination sensor (light sensor) of the display device, and operates the light emission of the above-described background LED based on the measured amount of ambient light. You can also control it. That is, the analysis server 1200 can measure the amount of surrounding light using the illuminance sensor (light sensor) of the display device and link the background LED emission.
분석 서버(1200)는 생체 데이터의 분석을 기반으로, 사용자의 최적의 수면 시간, 낮잠시간, 잠들기 시간(즉, 취침시간) 등을 도출할 수 있고, 이를 토대로 알람 기능을 제공할 수 있다.The analysis server 1200 can derive the user's optimal sleep time, nap time, and bedtime (i.e., bedtime) based on analysis of biometric data, and provide an alarm function based on this.
또한, 분석 서버(1200)는 수면 기법(일예로 신경성 피로 수면 모드, 호흡 훈련 등의 기법)과 관련된 최면, 호흡 관련 영상 등이 스마트 워치(1100) 또는 사용자 단말로 제공될 수 있도록 할 수 있다. 이때, 분석 서버(1200)는 신경성 피로 수면 모드로 제어시, 일예로 호흡 관련 영상을 제공할 때 해당 호흡 관련 영상을 550~600 nm 파장의 빛에 해당하는 석양 빛(주황색상)에 기반한 영상으로 재생성하고, 재생성된 호흡 관련 영상을 사용자 단말뿐만 아니라 스마트 워치(1100)에 함께(동시에) 제공하고, 이에 따라 스마트 워치 및 사용자 단말에서 재생성된 호흡 관련 영상의 재생이 이루어지도록 할 수 있다.In addition, the analysis server 1200 can provide hypnosis and breathing-related images related to sleep techniques (for example, techniques such as nervous fatigue sleep mode and breathing training) to the smart watch 1100 or the user terminal. At this time, when controlling the analysis server 1200 to the nervous fatigue sleep mode, for example, when providing a breathing-related image, the breathing-related image is converted into an image based on sunset light (orange color) corresponding to light with a wavelength of 550 to 600 nm. It is possible to reproduce and provide (simultaneously) the regenerated breathing-related images to the smart watch 1100 as well as the user terminal, thereby allowing the smart watch and the user terminal to play the regenerated breathing-related images.
또한, 분석 서버(1200)는 호흡 훈련 관련 영상(일예로 코로 숨 들이쉬기와 같은 호흡 훈련에 관한영상 콘텐츠)를 스마트 워치(1100)로 제공할 수 있는데, 이때 앞서 말한 바와 같이 해당 호흡 훈련 관련 영상 역시 550~600 nm 파장의 빛에 해당하는 석양 빛(주황색상)에 기반한 영상으로 재생성하여 스마트 워치(1100)로 제공할 수 있다.In addition, the analysis server 1200 may provide breathing training-related videos (e.g., video content related to breathing training, such as breathing in through the nose) to the smart watch 1100. In this case, as mentioned above, the breathing training-related video Also, an image based on sunset light (orange color) corresponding to light with a wavelength of 550 to 600 nm can be reproduced and provided to the smart watch (1100).
상술한 설명에 따르면, 분석 서버(1200)는 사용자에 대한 실시간 수면 코칭을 위하여, 인공지능 모델을 이용하여 기상시간을 산출하고 산출된 기상시간을 토대로 기상알람을 제공하는 기능, 블루라이트 차단모드의 ON/OFF 제어 기능, 블루라이트 증강 모드의 ON/OFF 제어 기능, 배경 라이트 LED를 제공하는 기능, 뇌파 유도 사운드를 제공하는 기능, 백업 알람 기능 등 다양한 기능을 사용자에게 제공할 수 있다.According to the above description, the analysis server 1200 calculates the wake-up time using an artificial intelligence model for real-time sleep coaching for the user, provides a wake-up alarm based on the calculated wake-up time, and provides a blue light blocking mode. Various functions can be provided to the user, such as ON/OFF control function, ON/OFF control function of blue light augmentation mode, function to provide background light LED, function to provide brain wave-induced sound, and backup alarm function.
이때, 분석 서버(1200)는 전술한 다양한 기능들을 제공함에 있어서, 이들 다양한 기능들을 모두 인공지능 모델에 기반하여 제공 가능하도록 마련되는 것이 아니라, 다양한 기능들 중 적어도 일부의 기능들(일예로 블루라이트 증강 모드의 ON/OFF 제어 기능, 백업 알람 기능 등)에 대해서는 인공지능에 국한될 필요 없이 인공지능 모델 없이도 사용자에게 제공 가능하도록 마련될 수 있다. 즉, 분석 서버(1200)에 의해 제공되는 다양한 기능들 중 적어도 일부의 기능들(일예로 블루라이트 증강 모드의 ON/OFF 제어 기능, 백업 알람 기능 등)은 인공지능에 국한되어 제공되는 것이 아니라, 인공지능 모델 없이도 인간(사용자)에 의한 설정에 의하여 혹은 일반적인 상황에서도 사용 가능하도록, 그 범위를 확장하여 전반적으로 사용 가능하게 적용(구현, 마련)될 수 있다.At this time, in providing the various functions described above, the analysis server 1200 is not prepared to provide all of these various functions based on an artificial intelligence model, but rather provides at least some of the various functions (for example, blue light Augmented mode ON/OFF control function, backup alarm function, etc.) do not need to be limited to artificial intelligence and can be provided to users without an artificial intelligence model. That is, at least some of the various functions provided by the analysis server 1200 (for example, the ON/OFF control function of the blue light augmentation mode, the backup alarm function, etc.) are not limited to artificial intelligence, but Even without an artificial intelligence model, it can be applied (implemented, prepared) for general use by expanding its scope so that it can be used by setting by a human (user) or in general situations.
이하에서는 상기에 자세히 설명된 내용을 기반으로, 본 발명의 동작 흐름을 간단히 살펴보기로 한다.Below, based on the details described above, we will briefly look at the operation flow of the present invention.
도 15는 본 발명의 일 실시예에 따른 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템의 신호 흐름을 나타낸 도면이다. 달리 말해, 도 15는 본 발명의 일 실시예에 따른 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 방법의 개략적인 동작 흐름을 나타낸 도면이다. Figure 15 is a diagram showing the signal flow of a real-time sleep health management service providing system using AI-based brain wave entrainment and autonomic nervous system control according to an embodiment of the present invention. In other words, Figure 15 is a diagram illustrating a schematic operational flow of a method for providing real-time sleep health management service using AI-based brain wave tuning and autonomic nervous system control according to an embodiment of the present invention.
이때, 도 15에 도시된 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 방법은 앞서 설명된 본 시스템(11)에 의하여 수행될 수 있다. 따라서, 이하 생략된 내용이라고 하더라도 본 시스템(11)에 대하여 설명된 내용은 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 방법에 대한 설명에도 동일하게 적용될 수 있다.At this time, the method of providing real-time sleep health management service using AI-based brain wave entrainment and autonomic nervous system control shown in FIG. 15 can be performed by the system 11 described above. Therefore, even if the content is omitted below, the content described about the system 11 can be equally applied to the explanation of the method of providing real-time sleep health management service using AI-based brain wave entrainment and autonomic nervous system control.
도 15를 참조하면, S150 단계에서는 스마트 워치(1100)에서 수면 중 사용자의 심박동, 움직임, 혈압을 포함하는 수면 생체 데이터를 수집하고, 수집된 수면 생체 데이터를 분석 서버(1200)로 전송한다. S155 단계에서는 분석 서버(1200)에서 수면 생체 데이터를 분석하여 S160 단계에서 최적 수면주기 정보 및 기상시간을 산출하고 산출된 최적 수면주기 정보 및 기상시간 정보를 스마트 워치(1100)로 전송한다. S165 단계에서는 수신한 기상시간에 따라 사용자에게 웨이크업 알람(기상알람)을 제공하고, S170 단계에서는 사용자 생체 정보를 수집한다. 여기서, 사용자 생체 정보는 사용자의 활동 중 생체 정보로서 심박수, 혈압, 움직임 여부 정보 등을 포함할 수 있다. S175 단계에서는 스마트 워치(1100)로부터 사용자 생체 정보를 수집하여 사용자의 기상 여부(즉, 기상상태인지 여부)를 확인한다. S175 단계에서 사용자가 기상한 것으로 확인(즉, 기상상태인 것으로 확인)되면, S150 단계로 다시 진입하고 S175 단계에서 사용자가 기상하지 않은 것으로 파악되는 경우에는 S180 단계로 진입한다. S180 단계에서는 스마트 워치(1100)에서 재 알람을 제공하여 사용자가 기상하도록 한다.Referring to FIG. 15 , in step S150, the smart watch 1100 collects sleep biometric data including the user's heart rate, movement, and blood pressure while sleeping, and transmits the collected sleep biometric data to the analysis server 1200. In step S155, the analysis server 1200 analyzes sleep biometric data, calculates optimal sleep cycle information and wake-up time in step S160, and transmits the calculated optimal sleep cycle information and wake-up time information to the smart watch 1100. In step S165, a wake-up alarm (wake-up alarm) is provided to the user according to the received wake-up time, and in step S170, the user's biometric information is collected. Here, the user biometric information is biometric information during the user's activities and may include heart rate, blood pressure, and movement information. In step S175, the user's biometric information is collected from the smart watch 1100 to check whether the user is awake (i.e., whether the user is awake). If it is confirmed that the user has woken up (i.e., is confirmed to be in a waking state) in step S175, step S150 is re-entered, and if it is determined that the user is not woken up in step S175, step S180 is entered. In step S180, the smart watch 1100 provides a re-alarm to help the user wake up.
본 발명에서는 일예로 S180 단계에서 재 알람 제공 후 다시 S175 단계로 진입하여 사용자의 기상 여부(즉, 기상상태 여부)를 재확인할 수 있으며, 사용자가 완전히 잠에서 깨어날 때까지 알람을 반복 제공(즉, S180 단계를 반복 수행)할 수 있다. 또한, 이때 사용자가 스마트 워치(1100)에서 기상알람 종료 아이콘을 터치하면, 알람 반복이 종료되도록 할 수 있다.In the present invention, for example, after providing a re-alarm in step S180, it is possible to re-check whether the user is awake (i.e., whether he is awake) by entering step S175 again, and repeatedly providing an alarm until the user fully wakes up (i.e. Step S180 can be repeated). Additionally, at this time, if the user touches the wake-up alarm end icon on the smart watch 1100, the alarm repetition can end.
상술한 설명에서, S150 단계 내지 S180 단계는 본 발명의 구현예에 따라서, 추가적인 단계들로 더 분할되거나, 더 적은 단계들로 조합될 수 있다. 또한, 일부 단계는 필요에 따라 생략될 수도 있고, 단계 간의 순서가 변경될 수도 있다.In the above description, steps S150 to S180 may be further divided into additional steps or combined into fewer steps, depending on the implementation of the present invention. Additionally, some steps may be omitted or the order between steps may be changed as needed.
도 16은 본 발명의 일 실시예에 따른 기상알람 제공 방법을 나타낸 도면이다.Figure 16 is a diagram showing a method of providing a wake-up alarm according to an embodiment of the present invention.
이때, 도 16에 도시된 기상알람 제공 방법은 앞서 설명된 분석 서버(1200)에 의하여 수행될 수 있다. 따라서, 이하 생략된 내용이라고 하더라도 분석 서버(1200)에 대하여 설명된 내용은 기상알람 제공 방법에 대한 설명에도 동일하게 적용될 수 있다.At this time, the method of providing a weather alarm shown in FIG. 16 may be performed by the analysis server 1200 described above. Therefore, even if the content is omitted below, the content described with respect to the analysis server 1200 can be equally applied to the explanation of the method of providing a weather alarm.
도 16을 참조하면, S161 단계에서는 분석 서버에서 스마트 워치를 이용해, 사용자의 움직임 유무(여부)를 분류한다. S162 단계에서는 사용자의 움직임이 없을 때의 심박수 데이터를 축적하고, 데이터 딥러닝 모델을 학습하여 사용자의 수면단계를 예측한다. S163단계에서는 사용자의 수면단계, 심박수 변화량을 고려해 램(REM) 수면단계 여부를 파악하고, S164 단계에서는 램(REM) 수면단계 이후 얕은 수면단계에서 스마트 워치(1100)를 통해 기상알람(기상알림)을 제공하도록 한다. 이때, S164 단계에서 분석 서버는, 일예로 사용자의 움직임과 심박수가 기상상태임을 확인할 때까지 알람을 반복 생성하여 제공(즉, 기상알람을 반복적으로 생성해 스마트 워치를 통해 제공)되도록 할 수 있다. Referring to FIG. 16, in step S161, the analysis server uses a smart watch to classify the presence (or absence) of the user's movement. In step S162, heart rate data is accumulated when the user does not move, and a deep learning data model is learned to predict the user's sleep stage. In step S163, the user's sleep stage and heart rate change are considered to determine whether the user is in the REM sleep stage, and in step S164, a wake-up alarm (wake-up notification) is sent through the smart watch (1100) in the light sleep stage after the REM sleep stage. shall be provided. At this time, in step S164, the analysis server may repeatedly generate and provide an alarm until, for example, it confirms that the user's movement and heart rate are in a waking state (that is, a waking-up alarm may be repeatedly generated and provided through a smart watch).
상술한 설명에서, S161 단계 내지 S165 단계는 본 발명의 구현예에 따라서, 추가적인 단계들로 더 분할되거나, 더 적은 단계들로 조합될 수 있다. 또한, 일부 단계는 필요에 따라 생략될 수도 있고, 단계 간의 순서가 변경될 수도 있다.In the above description, steps S161 to S165 may be further divided into additional steps or combined into fewer steps, depending on the implementation of the present invention. Additionally, some steps may be omitted or the order between steps may be changed as needed.
도 17은 본 발명의 일 실시예에 따른 사용자의 실제 수면시간을 반영한 기상시간 조정 방법을 나타낸 도면이다.Figure 17 is a diagram showing a method of adjusting the wake-up time reflecting the user's actual sleep time according to an embodiment of the present invention.
이때, 도 17에 도시된 사용자의 실제 수면시간을 반영한 기상시간 조정 방법은 앞서 설명된 분석 서버(1200)에 의하여 수행될 수 있다. 따라서, 이하 생략된 내용이라고 하더라도 분석 서버(1200)에 대하여 설명된 내용은 사용자의 실제 수면시간을 반영한 기상시간 조정 방법에 대한 설명에도 동일하게 적용될 수 있다.At this time, the method of adjusting the wake-up time that reflects the user's actual sleep time shown in FIG. 17 can be performed by the analysis server 1200 described above. Therefore, even if the content is omitted below, the content described with respect to the analysis server 1200 can be equally applied to the explanation of the method of adjusting the wake-up time to reflect the user's actual sleep time.
도 17을 참조하면, S1100 단계에서는 분석 서버가 스마트 워치로부터 생체 정보를 수집하고 S1200 단계에서는 생체 정보 분석을 통해, 사용자가 수면에 드는 수면 시작 시간을 파악한다. S1300 단계에서는 수면 시작 시간을 평소 패턴과 비교한다. S1400 단계에서는 비교 결과에 따라 반복되는 수면주기를 고려하여 기상시간을 조정한다.Referring to FIG. 17, in step S1100, the analysis server collects biometric information from the smart watch, and in step S1200, the sleep start time when the user goes to sleep is determined through biometric information analysis. In step S1300, the sleep start time is compared with the usual pattern. In step S1400, the wake-up time is adjusted by considering the repeated sleep cycle according to the comparison results.
예컨대, 평소 수면에 드는 시간보다 일찍 또는 늦게 잠들게 되면 수면시간이 달라지므로, S1400 단계에서 분석 서버는, 기상해야 하는 시점까지 수면할 수 있는 남은 시간을 고려해 예측되는 마지막 수면주기 중 얕은 수면단계에 사용자가 일어날 수 있도록 기상시간을 산출할 수 있으며, 이후, 산출된 기상시간을 스마트 워치로 전송하여 해당 시간에 사용자가 기상알람을 제공받을 수 있도록 할 수 있다.For example, if you fall asleep earlier or later than your usual sleep time, your sleep time will change, so in step S1400, the analysis server determines that the user is in the light sleep stage of the last sleep cycle predicted by considering the remaining time to sleep until the time to wake up. The wake-up time can be calculated so that the wake-up time can occur, and then the calculated wake-up time can be transmitted to the smart watch so that the user can receive a wake-up alarm at that time.
이때, 분석 서버(1200)는 스마트 워치를 이용해 사용자의 움직임이나 심박수 등 생체 정보를 수집하고 분석하는 것뿐만 아니라, 딥러닝을 통해 사용자별 수면 습관 데이터를 획득한다. 또한, 분석 서버(1200)는 획득한 수면 습관 데이터로부터 평균적으로 수면에 들어가는 실제 사용자의 수면 시작 시간과 비교하여 차이를 산출하고, 산출된 차이만큼의 시간을 계산해 마지막 수면주기 중 얕은 수면단계에 스마트 워치를 통해 사용자를 깨울 수 있도록 할 수 있다. 구체적으로, 사용자가 평소보다 1시간 늦게 잠들거나 일찍 잠에 든 경우, 분석 서버(1200)는 정해진 시간에 일어나는 것이 아니라 수면주기 그래프에 따라 1.5시간의 수면주기가 일정 횟수 반복된 이후의 시간 까지만 자고 깨어날 수 있도록 기상시간을 산출할 수 있다. 예컨대, 분석 서버(1200)는 수면주기가 3번 반복된 4.5 시간 또는 4회 반복된 6시간 이후에 사용자가 기상하도록 알람 시간을 산출할 수 있다. At this time, the analysis server 1200 not only collects and analyzes biometric information such as the user's movements and heart rate using a smart watch, but also acquires sleep habit data for each user through deep learning. In addition, the analysis server 1200 calculates the difference from the acquired sleep habit data by comparing it with the sleep start time of the actual user who goes to sleep on average, and calculates a time equal to the calculated difference to enter the smart sleep stage in the last sleep cycle. You can wake the user through the watch. Specifically, if the user falls asleep 1 hour later or earlier than usual, the analysis server 1200 does not wake up at a set time, but only sleeps until the time after the 1.5-hour sleep cycle is repeated a certain number of times according to the sleep cycle graph. You can calculate the wake-up time so you can wake up. For example, the analysis server 1200 may calculate an alarm time so that the user wakes up after 4.5 hours when the sleep cycle is repeated 3 times or 6 hours after the sleep cycle is repeated 4 times.
또한, 분석 서버(1200)는 머신러닝으로 학습한 수면 생체 데이터를 통해 잠자리에 누워있는 시간 대비 실제 잠든 시간인 수면효율이 일정 수준 이상인 취침시간과 기상시간을 산출하여, 산출된 취침시간과 기상시간을 알림 하도록 한다. 아울러, 분석 서버(1200)는 사용자의 수면 시작 시간이 평소 패턴과 일정 수준 이상 달라지는 경우, 평소 패턴과 비교해 부족한 수면시간을 산출하여 제공하고(알리고), 부족한 수면시간을 낮잠, 쪽잠으로 보충하도록 안내하는 보충 수면시간 안내 정보를 생성하여 제공할 수 있다. 예컨대, 하루 동안의 필수 수면시간이 8시간인데 사용자가 5시간 밖에 못잔 경우, 분석 서버(1200)는 사용자가 보충해야 하는 3시간 수면시간을 낮에 낮잠이나 쪽잠 방식으로 보충할 수 있도록 안내할 수 있다. In addition, the analysis server 1200 calculates bedtime and wake-up time at a certain level or higher in sleep efficiency, which is the actual sleeping time compared to the time lying in bed, through sleep biometric data learned through machine learning, and the calculated bedtime and wake-up time. Be notified. In addition, if the user's sleep start time differs from the user's usual pattern by a certain level or more, the analysis server 1200 calculates and provides (notifies) the insufficient sleep time compared to the usual pattern and guides the user to supplement the insufficient sleep time with a nap or nap. Supplementary sleep time guidance information can be created and provided. For example, if the required sleep time per day is 8 hours, but the user can only sleep 5 hours, the analysis server 1200 can guide the user to make up for the 3 hours of sleep time by taking a nap or napping during the day. there is.
또한, 분석 서버(1200)는 사용자가 낮잠을 취했을 때에도 스마트 워치를 이용해, 사용자의 수면 생체 데이터를 축적하여 분석하고, 수면주기를 고려해 사용자를 깨울 수 있도록 한다. 또한, 분석 서버(1200)는 낮잠의 경우에도 사용자가 완전히 일어났는지 여부를 생체 데이터 분석을 통해 파악하여 반복적인 기상알람을 제공할 수 있도록 한다. 또한, 분석 서버(1200)는 사용자가 평소 낮잠을 자는 경우, 평소 수면에 드는 시간보다 일찍 또는 늦게 잠들게 되면 수면시간이 달라지므로, 기상해야 하는 시점까지 남은 시간을 고려해 예측되는 마지막 수면주기 중 얕은 수면단계에 일어나도록 기상시간을 산출할 수 있다. In addition, the analysis server 1200 uses a smart watch to accumulate and analyze the user's sleep biometric data even when the user takes a nap, and wakes the user in consideration of the sleep cycle. Additionally, even in the case of a nap, the analysis server 1200 determines whether the user has completely woken up through biometric data analysis and provides a repetitive wake-up alarm. In addition, when the user usually takes a nap, the sleep time changes if the user falls asleep earlier or later than the usual sleep time, so the analysis server 1200 determines the light sleep during the last sleep cycle predicted by considering the time remaining until the time to wake up. The wake-up time can be calculated to wake up at the right time.
상술한 설명에서, S1100 단계 내지 S1400 단계는 본 발명의 구현예에 따라서, 추가적인 단계들로 더 분할되거나, 더 적은 단계들로 조합될 수 있다. 또한, 일부 단계는 필요에 따라 생략될 수도 있고, 단계 간의 순서가 변경될 수도 있다.In the above description, steps S1100 to S1400 may be further divided into additional steps or combined into fewer steps, depending on the implementation of the present invention. Additionally, some steps may be omitted or the order between steps may be changed as needed.
상술한 설명에 따르면, 본 발명은 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템(11) 및 방법을 제공함으로써, 수면시간, 취침시간, 기상시간, 낮잠 시간 등을 개인의 수면주기에 따라 맞춤 관리(코칭)할 수 있고, 수면주기를 이용해 사용자가 짧게 자더라도 개운하게 잘 수 있도록 할 수 있으며, 이를 통해, 사용자의 시간 효용성, 수면효율 및 휴식 효율을 높여 삶의 질을 향상시킬 수 있도록 한다. According to the above description, the present invention provides a real-time sleep health management service provision system (11) and method using AI-based brain wave entrainment and autonomic nervous system control, thereby providing individual sleep time, bedtime, wake-up time, and nap time. Customized management (coaching) can be done according to the sleep cycle, and the sleep cycle can be used to ensure that the user can sleep refreshed even if he or she sleeps briefly. Through this, the quality of life is improved by increasing the user's time utility, sleep efficiency, and rest efficiency. Make it possible to do it.
또한, 본 발명은 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템(11) 및 방법을 제공함으로써, 사용자가 수면 후 충분한 회복이 되지 않은 경우에는 필요한 램(REM) 수면시간을 사용자에게 알려주거나, 사용자 회복을 위한 다양한 서비스를 제공할 수 있는바, 이를 통해 사용자의 스트레스 및 피로 회복을 속도를 증진시킬 수 있다.In addition, the present invention provides a real-time sleep health management service provision system (11) and method using AI-based brain wave entrainment and autonomic nervous system control, thereby reducing the required REM sleep time when the user does not recover sufficiently after sleep. It can inform the user or provide various services for user recovery, thereby speeding up the user's recovery from stress and fatigue.
또한, 본 발명은 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템(11) 및 방법을 제공함으로써, 수면단계 판단에 최적화된 맞춤형 트레이닝 데이터 셋(Training Data Set)을 기반으로 딥러닝 뉴럴 네트워크에 기초한 기계학습을 수행하여 수면단계 판단 모델을 구현할 수 있고, 이를 토대로 수면단계 판단 모델로부터 산출되는 수면단계 예측 및 기상시간 산출 결과의 품질을 보다 향상시킬 수 있다.In addition, the present invention provides a real-time sleep health management service provision system (11) and method using AI-based brain wave entrainment and autonomic nervous system control, providing deep sleep based on a customized training data set optimized for sleep stage determination. A sleep stage judgment model can be implemented by performing machine learning based on a learning neural network, and based on this, the quality of sleep stage prediction and wake-up time calculation results calculated from the sleep stage judgment model can be further improved.
도 1 내지 도 17을 통해 설명된 일 실시예에 따른 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 방법은, 컴퓨터에 의해 실행되는 애플리케이션이나 프로그램 모듈과 같은 컴퓨터에 의해 실행가능한 명령어를 포함하는 기록 매체의 형태로도 구현될 수 있다. 컴퓨터 판독 가능 매체는 컴퓨터에 의해 액세스될 수 있는 임의의 가용 매체일 수 있고, 휘발성 및 비휘발성 매체, 분리형 및 비분리형 매체를 모두 포함한다. 또한, 컴퓨터 판독가능 매체는 컴퓨터 저장 매체를 모두 포함할 수 있다. 컴퓨터 저장 매체는 컴퓨터 판독가능 명령어, 데이터 구조, 프로그램 모듈 또는 기타 데이터와 같은 정보의 저장을 위한 임의의 방법 또는 기술로 구현된 휘발성 및 비휘발성, 분리형 및 비분리형 매체를 모두 포함한다.The method of providing a real-time sleep health management service using AI-based brain wave entrainment and autonomic nervous system control according to an embodiment described with reference to FIGS. 1 to 17 includes commands executable by a computer, such as an application or program module executed by a computer. It can also be implemented in the form of a recording medium containing. Computer-readable media can be any available media that can be accessed by a computer and includes both volatile and non-volatile media, removable and non-removable media. Additionally, computer-readable media may include all computer storage media. Computer storage media includes both volatile and non-volatile, removable and non-removable media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules or other data.
전술한 본 발명의 일 실시예에 따른 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 방법은, 단말기에 기본적으로 설치된 애플리케이션(이는 단말기에 기본적으로 탑재된 플랫폼이나 운영체제 등에 포함된 프로그램을 포함할 수 있음)에 의해 실행될 수 있고, 사용자가 애플리케이션 스토어 서버, 애플리케이션 또는 해당 서비스와 관련된 웹 서버 등의 애플리케이션 제공 서버를 통해 마스터 단말기에 직접 설치한 애플리케이션(즉, 프로그램)에 의해 실행될 수도 있다. 이러한 의미에서, 전술한 본 발명의 일 실시예에 따른 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 방법은 단말기에 기본적으로 설치되거나 사용자에 의해 직접 설치된 애플리케이션(즉, 프로그램)으로 구현되고 단말기에 등의 컴퓨터로 읽을 수 있는 기록매체에 기록될 수 있다.The method of providing a real-time sleep health management service using AI-based brain wave entrainment and autonomic nervous system control according to an embodiment of the present invention described above includes an application installed by default on a terminal (this is a program included in a platform or operating system, etc., which is basically installed on the terminal). may include), and may be executed by an application (i.e. program) installed directly on the master terminal by the user through an application providing server such as an application store server, an application, or a web server related to the service. . In this sense, the method of providing a real-time sleep health management service using AI-based brain wave tuning and autonomic nervous system control according to an embodiment of the present invention described above is an application (i.e., program) installed by default on the terminal or directly installed by the user. It may be implemented and recorded on a computer-readable recording medium such as a terminal.
전술한 본 발명의 설명은 예시를 위한 것이며, 본 발명이 속하는 기술분야의 통상의 지식을 가진 자는 본 발명의 기술적 사상이나 필수적인 특징을 변경하지 않고서 다른 구체적인 형태로 쉽게 변형이 가능하다는 것을 이해할 수 있을 것이다. 그러므로 이상에서 기술한 실시예들은 모든 면에서 예시적인 것이며 한정적이 아닌 것으로 이해해야만 한다. 예를 들어, 단일형으로 설명되어 있는 각 구성 요소는 분산되어 실시될 수도 있으며, 마찬가지로 분산된 것으로 설명되어 있는 구성 요소들도 결합된 형태로 실시될 수 있다.The description of the present invention described above is for illustrative purposes, and those skilled in the art will understand that the present invention can be easily modified into other specific forms without changing the technical idea or essential features of the present invention. will be. Therefore, the embodiments described above should be understood in all respects as illustrative and not restrictive. For example, each component described as unitary may be implemented in a distributed manner, and similarly, components described as distributed may also be implemented in a combined form.
본 발명의 범위는 상기 상세한 설명보다는 후술하는 특허청구범위에 의하여 나타내어지며, 특허청구범위의 의미 및 범위 그리고 그 균등 개념으로부터 도출되는 모든 변경 또는 변형된 형태가 본 발명의 범위에 포함되는 것으로 해석되어야 한다.The scope of the present invention is indicated by the claims described below rather than the detailed description above, and all changes or modified forms derived from the meaning and scope of the claims and their equivalent concepts should be construed as being included in the scope of the present invention. do.
발명의 실시를 위한 형태는 위의 발명의 실시를 위한 최선의 형태에서 함께 기술되었다.The form for carrying out the invention has been described together with the above best form for carrying out the invention.
본 발명은 빛과 사운드를 이용하여 뇌파동조를 이룰 수 있도록 하고, 빛과 호흡법을 이용하여 자율신경계를 조절하도록 함으로써, 취침시각 및 기상시각에 맞도록 뇌파 및 자율신경계를 조율하고, 기상시각에 맞춰 스피커를 통하여 세타파, 알파파 및 베타파를 동조하는 모노럴비트 또는 바이노럴비트의 음원을 재생하여 기상시의 뇌파를 유도하며, 기상시각에 블루라이트의 햇빛샤워를 제공하여 각성효과와 함께 세로토닌을 생성하도록 하고, 생성된 세로토닌이 취침시각에 멜라토닌으로 변함으로써 취침 및 숙면을 더욱 유도할 수 있으며, 수동으로 또는 스트레스 지수를 모니터링한 후 임계값을 초과하면 스트레스를 낮추기 위해 교대체감각자극을 사용자 단말 및 웨어러블 기기를 통하여 제공하고, 시상하부 이상활성으로 발생하는 군발두통을 일주기리듬을 정상화시킴으로써 완화시킴과 동시에 비침습성 미주신경자을 통하여 통증을 완화시킬 수 있어 산업상 이용가능성이 있다.The present invention uses light and sound to achieve brain wave synchronization, and uses light and breathing to control the autonomic nervous system, thereby adjusting the brain waves and autonomic nervous system to suit the bedtime and wake-up time, and to match the wake-up time. By playing a sound source of monaural beats or binaural beats that synchronize theta waves, alpha waves, and beta waves through a speaker, it induces brain waves when waking up, and provides an awakening effect and serotonin by providing a blue light sunlight shower at the time of waking up. By converting the generated serotonin into melatonin at bedtime, it is possible to further induce bedtime and sound sleep. Manually or after monitoring the stress index, if it exceeds the threshold, alternating somatosensory stimulation is applied to the user's terminal to reduce stress. and wearable devices, and has the potential for industrial use as it can relieve cluster headaches caused by abnormal activity of the hypothalamus by normalizing the circadian rhythm and simultaneously relieve pain through the non-invasive vagus nerve.
Claims (20)
- 취침시각 및 기상시각을 설정하고, 취침시각에 부교감신경계를 항진시키는 호흡법인 수면유도법을 출력하면서 멜라토닌 비억제 파장의 빛을 조사하며, 기상시각에 알파파 또는 베타파의 뇌파동조를 위하여 알파밴드(Alpha-Band) 또는 베타밴드(Beta-Band)의 플리커(Flicker)를 삽입하여 출력하고, 블루라이트를 출력하여 각성을 유도하는 사용자 단말; 및Set bedtime and wake-up time, output sleep induction, a breathing method that stimulates the parasympathetic nervous system at bedtime, irradiate light of a melatonin-inhibiting wavelength, and use alpha band (alpha band) to synchronize brain waves with alpha or beta waves at wake-up time. A user terminal that inserts and outputs an Alpha-Band or Beta-Band flicker and outputs blue light to induce awakening; and상기 취침시각 및 기상시각에 뇌파동조 및 자율신경계조절을 위한 빛, 사운드, 진동 및 호흡법에 대한 설정을 저장하는 데이터베이스화부, 상기 사용자 단말로부터 취침시각 및 기상시각을 설정받는 설정부, 상기 사용자 단말에서 설정한 취침시각 및 기상시각에 상기 뇌파동조 및 자율신경계조절을 위한 빛, 사운드, 진동 및 호흡법 중 적어도 하나를 출력하도록 하는 제어부를 포함하는 관리 서비스 제공 서버;A database unit that stores settings for light, sound, vibration, and breathing methods for brain wave entrainment and autonomic nervous system control at the bedtime and wake-up time, a setting unit that receives settings for bedtime and wake-up time from the user terminal, and a setting unit that receives settings for bedtime and wake-up time from the user terminal. a management service providing server including a control unit configured to output at least one of light, sound, vibration, and breathing for brain wave tuning and autonomic nervous system control at set bedtime and wake-up time;를 포함하는 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템.A real-time sleep health management service provision system using AI-based brain wave entrainment and autonomic nervous system control, including.
- 제 1 항에 있어서,According to claim 1,상기 멜라토닌 비억제 파장의 빛은,The light of the melatonin non-suppressing wavelength is,적색 및 황색의 빛을 포함하는 것을 특징으로 하는 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템.A real-time sleep health management service provision system using AI-based brain wave entrainment and autonomic nervous system control, characterized by including red and yellow light.
- 제 1 항에 있어서,According to claim 1,상기 관리 서비스 제공 서버는,The management service providing server is,상기 기상시각으로부터 기 설정된 시간 이전부터 상기 사용자 단말의 스피커, 이어폰 및 헤드셋 중 어느 하나를 통하여 세타파, 알파파 및 베타파의 뇌파동조를 위하여 모노럴비트(Monaural Beat) 또는 바이노럴비트(Binaural Beats)의 소리를 출력하는 기상뇌파유도부;Monaural Beats or Binaural Beats for brain wave entrainment of theta waves, alpha waves, and beta waves through any one of the speakers, earphones, and headsets of the user terminal from before the preset time from the wake-up time. A wake-up brain wave induction unit that outputs sounds;를 더 포함하는 것을 특징으로 하는 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템.A real-time sleep health management service provision system using AI-based brain wave entrainment and autonomic nervous system control, further comprising:
- 제 1 항에 있어서,According to claim 1,상기 관리 서비스 제공 서버는,The management service providing server is,상기 취침시각으로부터 상기 사용자 단말과 연동된 웨어러블 기기에서 수면단계를 모니터링하고, 수면단계별로 기 설정된 모노럴비트 또는 바이노럴비트를 가청 주파수 또는 비가청 주파수로 출력하는 수면단계별코칭부;A sleep stage-specific coaching unit that monitors sleep stages in a wearable device linked to the user terminal from the bedtime and outputs monaural beats or binaural beats preset for each sleep stage at an audible or inaudible frequency;를 더 포함하고,It further includes,상기 사용자 단말과 이어폰 또는 헤드폰이 연동된 경우에는 바이노럴비트를 출력하거나, 상기 사용자 단말과 스피커와 연결된 경우에는 모노럴비트를 출력하는 것을 특징으로 하는 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템.Real-time sleep using AI-based brain wave entrainment and autonomic nervous system control, characterized in that binaural beats are output when the user terminal and earphones or headphones are linked, or monaural beats are output when the user terminal is connected to a speaker. Health care service delivery system.
- 제 4 항에 있어서,According to claim 4,상기 관리 서비스 제공 서버는,The management service providing server is,상기 수면단계를 파악하기 위하여 상기 사용자 단말 또는 상기 사용자 단말과 연동된 상기 웨어러블 기기로부터 수집된 수집 데이터를 기 구축된 인공지능 알고리즘에 질의로 입력하여 상기 수면단계를 파악하는 인공지능부;An artificial intelligence unit that determines the sleep stage by inputting collected data collected from the user terminal or the wearable device linked with the user terminal as a query to a pre-built artificial intelligence algorithm to determine the sleep stage;를 더 포함하는 것을 특징으로 하는 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템.A real-time sleep health management service provision system using AI-based brain wave entrainment and autonomic nervous system control, further comprising:
- 제 4 항에 있어서,According to claim 4,상기 관리 서비스 제공 서버는,The management service providing server is,상기 사용자 단말에서 불면증으로 등록하거나 불면증으로 등록하지 않은 경우, 수동설정으로 또는 수면단계별로 알파파, 세타파 및 델타파를 동조시키도록 모노럴비트 또는 바이노럴비트를 제공하고, REM 수면을 유도하기 위한 깊은 수면단계 후에 뇌파동조를 중지하며, 상기 사용자 단말에서 수면과다증으로 등록하거나 수면과다증으로 등록하지 않은 경우 수동설정으로 또는 세타파, 알파파 및 베타파의 순서로 동조시키도록 상기 모노럴비트 또는 바이노럴비트를 제공하는 불면과다완화부;If the user terminal registers as insomnia or does not register as insomnia, monaural beats or binaural beats are provided to synchronize alpha waves, theta waves, and delta waves by manual setting or by sleep stage, and to induce REM sleep. After the deep sleep stage, brain wave synchronization is stopped, and if the user terminal registers as hypersomnia or does not register as hypersomnia, the monaural beat or binaural beat is used to synchronize by manual setting or in the order of theta waves, alpha waves, and beta waves. Insomnia and Hyperactivity Relief Department, which provides beats;를 더 포함하고,It further includes,상기 사용자 단말과 이어폰 또는 헤드폰이 연동된 경우에는 바이노럴비트를 출력하거나, 상기 사용자 단말과 스피커와 연결된 경우에는 모노럴비트를 출력하는 것을 특징으로 하는 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템.Real-time sleep using AI-based brain wave entrainment and autonomic nervous system control, characterized in that binaural beats are output when the user terminal and earphones or headphones are linked, or monaural beats are output when the user terminal is connected to a speaker. Health care service delivery system.
- 제 6 항에 있어서,According to claim 6,상기 불면과다완화부는,The insomnia hyperalleviation department,델타파 뇌파 동조를 이용하여 첫 수면 사이클의 깊은 수면단계 시간을 증가시키고, 수면단계를 모니터링하여 REM 수면 단계에 기상시켜 REM 수면과잉을 비활성화시키는 것을 특징으로 하는 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템.Using AI-based brain wave entrainment and autonomic nervous system control, which uses delta wave brain wave entrainment to increase the time of the deep sleep stage of the first sleep cycle, monitors the sleep stage and wakes up in the REM sleep stage to deactivate REM sleep excess. Real-time sleep health management service provision system.
- 제 1 항에 있어서,According to claim 1,상기 관리 서비스 제공 서버는,The management service providing server is,상기 기상시각에 맞추어 블루라이트인 청색의 빛을 제공하는 햇빛샤워부;A sunlight shower unit that provides blue light according to the waking time;를 더 포함하는 것을 특징으로 하는 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템.A real-time sleep health management service provision system using AI-based brain wave entrainment and autonomic nervous system control, further comprising:
- 제 8 항에 있어서,According to claim 8,상기 관리 서비스 제공 서버는,The management service providing server is,상기 청색의 빛을 출력할 시간을 상기 기상시각 이외에 상기 사용자 단말의 사용자 데이터로 파악된 일주기리듬에 기초하여 상기 청색의 빛을 조사하도록 하는 일주기리듬맞춤부;A circadian rhythm adjusting unit that irradiates the blue light based on the circadian rhythm determined by user data of the user terminal in addition to the wake-up time for the time to output the blue light;를 더 포함하는 것을 특징으로 하는 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템.A real-time sleep health management service provision system using AI-based brain wave entrainment and autonomic nervous system control, further comprising:
- 제 9 항에 있어서,According to clause 9,상기 관리 서비스 제공 서버는,The management service providing server is,상기 일주기리듬의 장애가 모니터링된 경우, 미주신경자극기(NonInvasive Vagus Nerve Stimulation)에 대응하는 진동을 상기 사용자 단말에서 출력하도록 하거나 상기 미주신경자극기와 연동시켜 상기 미주신경자극기를 구동시키는 미주신경자극부;When the circadian rhythm disorder is monitored, a vagus nerve stimulator that outputs vibration corresponding to a vagus nerve stimulator (NonInvasive Vagus Nerve Stimulation) from the user terminal or operates the vagus nerve stimulator in conjunction with the vagus nerve stimulator;를 더 포함하는 것을 특징으로 하는 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템.A real-time sleep health management service provision system using AI-based brain wave entrainment and autonomic nervous system control, further comprising:
- 제 1 항에 있어서,According to claim 1,상기 관리 서비스 제공 서버는,The management service providing server is,상기 사용자 단말과 연동된 웨어러블 기기로부터 기 설정된 스트레스 지수를 초과하는 스트레스가 모니터링되거나 수동으로 설정된 경우, 상기 사용자 단말을 사용자의 한 쪽 손 또는 손목에, 상기 웨어러블 기기를 상기 사용자의 다른 쪽 손 또는 손목에 위치시키도록 한 후, 기 설정된 주파수를 가지는 진동을 출력하도록 하는 교대체감각자극부;When stress exceeding a preset stress index is monitored or manually set by a wearable device linked to the user terminal, the user terminal is placed on one hand or wrist of the user, and the wearable device is placed on the other hand or wrist of the user. An alternating somatosensory stimulation unit that is positioned in and then outputs vibration having a preset frequency;를 더 포함하고,It further includes,상기 진동의 주파수, 강도, 지속시간 및 횟수를 상기 사용자 단말에서 증감하도록 설정하는 것을 특징으로 하는 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템.A real-time sleep health management service providing system using AI-based brain wave entrainment and autonomic nervous system control, characterized in that the frequency, intensity, duration, and number of vibrations are set to increase or decrease in the user terminal.
- 수면 중 사용자의 수면 생체 데이터를 수집하는 스마트 워치; 및A smartwatch that collects the user's sleep biometric data while sleeping; and상기 스마트 워치로부터 수집한 수면 생체 데이터를 축적하고, 축적된 수면 생체 데이터를 인공지능 모델로 학습하고 분석함으로써 사용자별 수면 패턴 및 최적 수면주기 정보를 파악하고, 상기 최적 수면주기 정보에 기반한 기상시간을 산출하는 분석 서버를 포함하고,By accumulating sleep biometric data collected from the smart watch, learning and analyzing the accumulated sleep biometric data with an artificial intelligence model, sleep patterns and optimal sleep cycle information for each user are identified, and wake-up time based on the optimal sleep cycle information is determined. Includes an analysis server that calculates,상기 스마트 워치는, 상기 분석 서버로부터 상기 산출된 기상시간의 기상시간 정보를 전달받아 상기 기상시간에 기상알람을 제공하는 것을 특징으로 하는 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템.The smart watch provides a real-time sleep health management service using AI-based brain wave entrainment and autonomic nervous system regulation, characterized by receiving wake-up time information of the calculated wake-up time from the analysis server and providing a wake-up alarm at the wake-up time. system.
- 제 12 항에 있어서,According to claim 12,상기 분석 서버는, 상기 기상시간으로서, 상기 산출된 최적 수면주기 정보에 따라 램(REM) 수면단계 이후 얕은 수면단계에서 사용자를 깨우는 기상시간을 산출하는 것을 특징으로 하는 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템.The analysis server is an AI-based brain wave entrainment and autonomic nervous system control, characterized in that, as the wake-up time, it calculates a wake-up time to wake the user from a light sleep stage after the REM sleep stage according to the calculated optimal sleep cycle information. A real-time sleep health management service provision system using .
- 제 12 항에 있어서,According to claim 12,상기 분석 서버는, 상기 스마트 워치로부터 실시간 수신되는 생체 정보를 분석하여 사용자가 기상상태인지 확인하고, 기상상태가 아닌 것으로 확인되는 경우 일정 시간 간격으로 기상알람을 반복 생성하여 제공되도록 하는 것을 특징으로 하는 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템.The analysis server analyzes biometric information received in real time from the smart watch to check whether the user is in a weather state, and when it is confirmed that the user is not in a weather state, it repeatedly generates and provides a wake-up alarm at regular time intervals. A real-time sleep health management service provision system using AI-based brain wave entrainment and autonomic nervous system control.
- 제 14 항에 있어서,According to claim 14,상기 분석 서버는, 사용자가 기상상태인 것으로 확인되면, 기 설정되어 있는 목표 행동을 확인한후 상기 스마트 워치에서 상기 확인된 목표 행동에 대응하는 생체 데이터가 획득될 때까지 상기 기상알람을 반복 생성하여 제공되도록 하는 것을 특징으로 하는 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템.When the analysis server determines that the user is in a good state of waking up, the analysis server checks the preset target behavior and then repeatedly generates and provides the wake-up alarm until biometric data corresponding to the confirmed target behavior is obtained from the smart watch. A real-time sleep health management service provision system using AI-based brain wave entrainment and autonomic nervous system control, characterized by
- 제 12 항에 있어서,According to claim 12,상기 분석 서버는, 사용자의 수면 시작 시간이 평소 패턴과 일정 수준 이상 달라지는 경우, 평소 패턴과 비교해 부족한 수면시간을 산출하여 제공하는 것을 특징으로 하는 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템.The analysis server provides real-time sleep health management using AI-based brain wave entrainment and autonomic nervous system regulation, characterized by calculating and providing insufficient sleep time compared to the usual pattern when the user's sleep start time differs from the usual pattern by more than a certain level. Service delivery system.
- 제 12 항에 있어서,According to claim 12,상기 분석 서버는, 상기 기상시간 정보를 고려하여 상기 스마트 워치에 대한 블루라이트 차단 모 드와 블루라이트 증강 모드의 ON/OFF를 제어하는 것을 특징으로 하는 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템.The analysis server controls ON/OFF of the blue light blocking mode and blue light enhancement mode for the smart watch in consideration of the wake-up time information. Real-time sleep using AI-based brain wave entrainment and autonomic nervous system control. Health care service delivery system.
- 제 12 항에 있어서,According to claim 12,상기 분석 서버는,The analysis server is,상기 스마트 워치 및 상기 사용자가 소지한 사용자 단말의 일영역에 설치 가능한 배경 라이트 LED를 제공하되,Provide a background light LED that can be installed in one area of the smart watch and the user terminal owned by the user,상기 배경 라이트 LED는 근시와 난시 예방이 가능한 형태로 마련되고, 상기 스마트 워치와 상기 사용자 단말을 포함하여 사용자에 의해 기 등록된 디스플레이 디바이스에 설치 가능하게 마련되며,The background light LED is provided in a form capable of preventing myopia and astigmatism, and is provided to be installed on display devices pre-registered by the user, including the smart watch and the user terminal,상기 디스플레이 디바이스에 설치되는 배경 라이트 LED와 상기 디스플레이 디바이스의 후면에 마련된 플래시 라이트 LED는, 상기 분석 서버에 의한 제어 또는 사용자 입력에 기반한 제어에 의해 사용, 설정 및 작동이 가능하도록 마련되는 것을 특징으로 하는 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템.The background light LED installed in the display device and the flash light LED provided on the rear of the display device are provided to enable use, setting, and operation by control by the analysis server or control based on user input. A real-time sleep health management service provision system using AI-based brain wave entrainment and autonomic nervous system control.
- 제 12 항에 있어서,According to claim 12,상기 분석 서버는, 상기 스마트 워치로부터 실시간 수신되는 생체 정보를 이용하여 분석된 사용자의 수면단계를 고려하여, 사용자의 수면 유도를 위해 복수의 뇌파 유도 사운드 중 적어도 하나의 뇌파 유도 사운드를 제공하는 것을 특징으로 하는 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템.The analysis server is characterized in that it provides at least one brain wave-induced sound among a plurality of brain wave-induced sounds to induce sleep of the user in consideration of the user's sleep stage analyzed using biometric information received in real time from the smart watch. A real-time sleep health management service provision system using AI-based brain wave entrainment and autonomic nervous system control.
- 제 12 항에 있어서,According to claim 12,상기 분석 서버는,The analysis server is,상기 기상시간에 상기 스마트 워치에서 기상알람이 제공되도록 제어하되,Control the wake-up alarm to be provided from the smart watch at the wake-up time,상기 기상시간에 상기 스마트 워치의 전원이 OFF 상태인 것으로 감지되는 경우, 사용자가 소지한 적어도 하나의 사용자 단말 중 전원이 ON 상태에 있는 사용자 단말에서 기상알람이 제공되도록 제어하는 것을 특징으로 하는 AI 기반 뇌파동조 및 자율신경계조절을 이용한 실시간 수면 건강 관리 서비스 제공 시스템.When it is detected that the power of the smart watch is OFF at the wake-up time, an AI-based device is controlled to provide a wake-up alarm from at least one user terminal owned by the user whose power is ON. A real-time sleep health management service provision system using brain wave entrainment and autonomic nervous system control.
Applications Claiming Priority (6)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR20220116608 | 2022-09-15 | ||
| KR10-2022-0116608 | 2022-09-15 | ||
| KR10-2022-0172430 | 2022-12-12 | ||
| KR1020220172430A KR20240038545A (en) | 2022-09-15 | 2022-12-12 | Real-time sleep coaching system and method using artificial intelligence |
| KR1020230103492A KR102643867B1 (en) | 2023-08-08 | 2023-08-08 | Sleep health care system using artificial intelligence based brainwave entrainment and autonomic nervous system control |
| KR10-2023-0103492 | 2023-08-08 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2024058488A1 true WO2024058488A1 (en) | 2024-03-21 |
Family
ID=90275309
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/KR2023/013327 WO2024058488A1 (en) | 2022-09-15 | 2023-09-06 | System for providing real-time sleep health management service by using ai-based brain wave synchronization and autonomic nervous system control |
Country Status (1)
| Country | Link |
|---|---|
| WO (1) | WO2024058488A1 (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20170123065A (en) * | 2016-04-28 | 2017-11-07 | 국민대학교산학협력단 | Illumination sensor and illumination control device using the same |
| KR20180026187A (en) * | 2016-09-02 | 2018-03-12 | 미미라이팅주식회사 | Lighting apparatus having wake-up and sleep inducing function |
| US20180110960A1 (en) * | 2013-03-15 | 2018-04-26 | Youngblood Ip Holdings, Llc | Stress reduction and sleep promotion system |
| KR20190119236A (en) * | 2018-04-05 | 2019-10-22 | 동명대학교산학협력단 | Personalized personal artificial intelligence information providing system and method using body and habit information |
| KR20200031483A (en) * | 2018-09-14 | 2020-03-24 | 유남구 | Schedule Recommendation System and Device Using Sleep Assist Device |
| KR20220038572A (en) * | 2020-09-20 | 2022-03-29 | 아펙스 뉴로 홀딩스 아이엔씨. | Devices and mehtods for using mechanical affective touch therapy to reduce stress, anxiety and depression |
-
2023
- 2023-09-06 WO PCT/KR2023/013327 patent/WO2024058488A1/en unknown
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180110960A1 (en) * | 2013-03-15 | 2018-04-26 | Youngblood Ip Holdings, Llc | Stress reduction and sleep promotion system |
| KR20170123065A (en) * | 2016-04-28 | 2017-11-07 | 국민대학교산학협력단 | Illumination sensor and illumination control device using the same |
| KR20180026187A (en) * | 2016-09-02 | 2018-03-12 | 미미라이팅주식회사 | Lighting apparatus having wake-up and sleep inducing function |
| KR20190119236A (en) * | 2018-04-05 | 2019-10-22 | 동명대학교산학협력단 | Personalized personal artificial intelligence information providing system and method using body and habit information |
| KR20200031483A (en) * | 2018-09-14 | 2020-03-24 | 유남구 | Schedule Recommendation System and Device Using Sleep Assist Device |
| KR20220038572A (en) * | 2020-09-20 | 2022-03-29 | 아펙스 뉴로 홀딩스 아이엔씨. | Devices and mehtods for using mechanical affective touch therapy to reduce stress, anxiety and depression |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113631212B (en) | System and method for wave generation for percutaneous vibration | |
| CN110049714B (en) | System and method for promoting wakefulness | |
| US20220362095A1 (en) | Systems and methods of transcutaneous vibration for sexual arousal | |
| US20190231256A1 (en) | Apparatus and associated methods for adjusting a user's sleep | |
| CN107466241B (en) | Providing sensory stimulation based on slow wave cycles | |
| JP2016539758A (en) | MULTIPHASE SLEEP MANAGEMENT SYSTEM, ITS OPERATION METHOD, SLEEP ANALYSIS DEVICE, CURRENT SLEEP PHASE CLASSIFICATION METHOD, MULTIPHASE SLEEP MANAGEMENT SYSTEM, AND USE OF SLEEP ANALYSIS DEVICE IN MULTIPHASE SLEEP MANAGEMENT | |
| CN111372639B (en) | System for delivering sensory stimuli to a user to enhance cognitive domains in the user | |
| JP7479150B2 (en) | SYSTEM AND METHOD FOR DETERMINING WHETHER A SUBJECT IS LIKELY TO BE DISTURBED BY THERAPEUTIC LEVELS OF STIMULATION DURING A SLEEP SESSION - Patent application | |
| WO2020153584A1 (en) | Electronic device and sleep induction method thereof | |
| WO2024058488A1 (en) | System for providing real-time sleep health management service by using ai-based brain wave synchronization and autonomic nervous system control | |
| KR102643867B1 (en) | Sleep health care system using artificial intelligence based brainwave entrainment and autonomic nervous system control | |
| KR102729208B1 (en) | Electronic apparatus and method of inducing sleep | |
| WO2024128478A1 (en) | Ai-based affective disorder digital diagnosis system | |
| KR102645578B1 (en) | Cognitive function improving service artificial intelligence based brainwave entrainment and light therapy | |
| WO2024059191A2 (en) | Systems and methods of temperature and visual stimulation patterns | |
| WO2023229598A1 (en) | Systems and methods of transcutaneous vibration | |
| WO2019245322A1 (en) | Stimulus generation system | |
| KR20230099805A (en) | Wake-up alarm setting and sleep induction device using EEG analysis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 23865786 Country of ref document: EP Kind code of ref document: A1 |