WO2012063308A1 - 睡眠状態推定装置 - Google Patents
睡眠状態推定装置 Download PDFInfo
- Publication number
- WO2012063308A1 WO2012063308A1 PCT/JP2010/069847 JP2010069847W WO2012063308A1 WO 2012063308 A1 WO2012063308 A1 WO 2012063308A1 JP 2010069847 W JP2010069847 W JP 2010069847W WO 2012063308 A1 WO2012063308 A1 WO 2012063308A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- subject
- state estimation
- sleep
- waveform
- state
- Prior art date
Links
- 230000007958 sleep Effects 0.000 title claims abstract description 109
- 230000029058 respiratory gaseous exchange Effects 0.000 claims abstract description 173
- 230000000241 respiratory effect Effects 0.000 claims abstract description 88
- 230000033001 locomotion Effects 0.000 claims abstract description 61
- 230000003187 abdominal effect Effects 0.000 claims description 36
- 230000001133 acceleration Effects 0.000 claims description 22
- 238000000034 method Methods 0.000 claims description 22
- 230000006399 behavior Effects 0.000 claims description 17
- 230000008667 sleep stage Effects 0.000 abstract description 29
- 238000012545 processing Methods 0.000 abstract description 7
- 210000000133 brain stem Anatomy 0.000 abstract description 3
- 210000000653 nervous system Anatomy 0.000 abstract 1
- 210000000038 chest Anatomy 0.000 description 31
- 238000001514 detection method Methods 0.000 description 13
- 230000005856 abnormality Effects 0.000 description 5
- 206010062519 Poor quality sleep Diseases 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 4
- 210000001015 abdomen Anatomy 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 210000000467 autonomic pathway Anatomy 0.000 description 2
- 239000006185 dispersion Substances 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 208000019901 Anxiety disease Diseases 0.000 description 1
- 210000000683 abdominal cavity Anatomy 0.000 description 1
- 230000036506 anxiety Effects 0.000 description 1
- 210000000481 breast Anatomy 0.000 description 1
- 210000001217 buttock Anatomy 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 210000002414 leg Anatomy 0.000 description 1
- 230000006996 mental state Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000017311 musculoskeletal movement, spinal reflex action Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 201000002859 sleep apnea Diseases 0.000 description 1
- 238000011282 treatment Methods 0.000 description 1
- 230000021542 voluntary musculoskeletal movement Effects 0.000 description 1
- 230000002618 waking effect 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
- A61B5/48—Other medical applications
- A61B5/4806—Sleep evaluation
- A61B5/4809—Sleep detection, i.e. determining whether a subject is asleep or not
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/08—Measuring devices for evaluating the respiratory organs
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/16—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state
- A61B5/18—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state for vehicle drivers or machine operators
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
Definitions
- the present invention relates to a sleep state estimation device, and more particularly to a sleep state estimation device that estimates a subject's state.
- Patent Literature 1 includes an in-bed determination unit, and the in-bed determination unit determines that the person is in bed when the sleeping person's heart rate exceeds the in-bed determination threshold, and falls below the in-bed determination threshold.
- An apparatus that is determined to be out of bed is disclosed. This presence / absence determination is stored in the memory.
- the determination by the presence determination unit is regarded as an erroneous determination.
- the above-described techniques it is possible to estimate only the presence of the bed and the leaving of the bed with high accuracy.
- the kurtosis of the frequency distribution of the heart rate which is a statistic of a certain section, is used to check whether the user is in bed or out of bed by collating with a database.
- body movements that occurred in a short time cannot be judged from the statistics.
- the above technique cannot distinguish between involuntary and voluntary movements.
- the above technique cannot identify the type of body movement, and cannot distinguish between intentional breathing such as conversation and deep breathing.
- the above technique when the body motion is continuously performed, for example, the motion of the subject unconsciously shaking the body cannot be distinguished from the normal body motion.
- the present invention has been made in consideration of such circumstances, and its purpose is to make it easier to classify the subject's body movements in more detail, and to target subjects such as sleep depth and body movements. It is in providing the sleep state estimation apparatus which implement
- the present invention is a sleep state estimation device including a state estimation unit that estimates the state of a subject based on the identity of the respiratory waveform for each period as a feature amount of the subject's respiratory waveform.
- a state estimation unit estimates a subject's state based on the identity for every period of a subject's respiration waveform as a feature-value of a respiration waveform. Because the subject's breathing fluctuations can be controlled at will, the subject's body movements can be further detailed by estimating the subject's state based on the identity of the subject's breathing waveform for each period. And the accuracy of estimating the state of the subject such as the sleep depth and body movement can be more easily improved.
- the state estimation unit as the identity of the subject's respiratory waveform for each cycle, the autocorrelation that is the identity of the waveform obtained by shifting the respiratory waveform by an arbitrary time and the original respiratory waveform, and the respiratory waveform
- the state of the subject can be estimated based on at least one of the reproducibility that is the fluctuation of the lowest value for each period.
- the state estimation unit as the identity of the subject's breathing waveform for each cycle, autocorrelation that is the identity of the waveform obtained by shifting the breathing waveform by an arbitrary time and the original breathing waveform,
- the state of the subject is estimated based on at least one of reproducibility, which is the fluctuation of the minimum value for each cycle of the respiratory waveform.
- the state estimation unit can estimate the state of the subject based on the period, amplitude, autocorrelation and reproducibility of the respiratory waveform as the feature quantity of the subject's respiratory waveform.
- the state estimation unit estimates the state of the subject based on the period, amplitude, autocorrelation and reproducibility of the respiratory waveform as the feature quantity of the subject's respiratory waveform.
- the state of the subject is estimated by combining the four indicators of the period, amplitude, autocorrelation and reproducibility of the respiratory waveform, so the accuracy of estimating the state of the subject such as sleep depth and body movement Can be further improved.
- the state estimation unit detects that the amplitude, autocorrelation and reproducibility of the respiration waveform are fluctuating compared to the normal time of the subject, the state estimation unit It can be assumed that there is.
- the state estimation unit detects that the amplitude, autocorrelation, and reproducibility of the respiration waveform are fluctuating as compared to the normal time of the subject, the subject is seated again. It is estimated that The present inventor has obtained knowledge that when the subject is in a re-sitting state, fluctuations in the amplitude, autocorrelation and reproducibility of the subject's respiratory waveform are observed. For this reason, if it is detected that the amplitude, autocorrelation and reproducibility of the respiration waveform are fluctuating compared to the normal time of the subject, it is assumed that the subject is sitting again Thus, it is possible to accurately estimate the state in which the subject is sitting back.
- the state estimation unit detects that the amplitude and reproducibility of the respiration waveform are fluctuating compared to the normal time of the subject, the state is that the subject is reaching out upwards. It can be estimated.
- the state estimation unit detects that the amplitude and reproducibility of the respiration waveform are fluctuating compared to the normal time of the subject, the subject is reaching out upwards. Presumed to be in a state.
- the present inventor has obtained knowledge that when the subject is in a state of reaching his hand upward, fluctuations in the amplitude and reproducibility of the subject's respiratory waveform are observed. For this reason, when it is detected that the amplitude and reproducibility of the respiration waveform are fluctuating compared to the normal time of the subject, by estimating that the subject is reaching out upward It is possible to accurately estimate the state in which the subject is reaching up.
- the state estimation unit detects that the autocorrelation and reproducibility of the respiration waveform are fluctuating as compared with the normal time of the subject, the state is that the subject is in conversation. It can be estimated.
- the state estimation unit detects that the autocorrelation and reproducibility of the respiration waveform are fluctuating compared to the normal time of the subject, the subject is talking. Presumed to be in a state.
- the present inventor has obtained the knowledge that when the subject is in a conversation state, the autocorrelation and reproducibility of the subject's respiratory waveform change. For this reason, if it is detected that the autocorrelation and reproducibility of the respiratory waveform are fluctuating compared to the normal time of the subject, by estimating that the subject is talking The state in which the subject is talking can be accurately estimated.
- the state estimation unit estimates that the subject is taking a deep breath when the period and amplitude of the respiratory waveform are detected to be changing compared to the subject's normal time. And can.
- the state estimation unit is in a state where the subject is taking a deep breath when it detects that the period and amplitude of the respiratory waveform are fluctuating compared to the normal time of the subject.
- the present inventor has obtained knowledge that when the subject is taking a deep breath, fluctuations are observed in the period and amplitude of the breathing waveform of the subject. For this reason, when it is detected that the period and amplitude of the respiration waveform are fluctuating compared to the normal time of the subject, the subject is estimated by assuming that the subject is taking a deep breath. Can accurately estimate the state of taking a deep breath.
- the state estimation unit can estimate the state of the subject by comparing the feature quantity of the respiratory waveform with the threshold value set for each feature quantity.
- the state estimation unit estimates the target person's state by comparing the feature value of the respiratory waveform with the threshold value set for each feature value. It is possible to estimate the state of the target person.
- the state estimation unit can set a threshold value for each feature amount for each target person.
- the state estimation unit sets a threshold value for each feature amount for each target person. For this reason, it becomes possible to estimate a subject's state corresponding to constitution and taste for every subject.
- the state estimation unit can estimate the state of the subject who has boarded the vehicle, and can estimate the state of the subject while discriminating the vehicle behavior of the vehicle and the body movement of the subject based on the acceleration of the vehicle. .
- the state estimation unit estimates the state of the subject who has boarded the vehicle, and estimates the state of the subject while discriminating the vehicle behavior of the vehicle and the body movement of the subject based on the acceleration of the vehicle. To do. For this reason, it is possible to accurately estimate the state of the subject who has boarded the vehicle while discriminating the vehicle behavior and the body movement.
- the state estimation unit can estimate the sleep depth of the subject by determining whether the subject's breathing method is abdominal breathing or chest breathing from the breathing waveform of the subject.
- the state estimation unit estimates the sleep depth of the subject by determining whether the subject's breathing method is abdominal breathing or chest breathing from the breathing waveform of the subject. Whether the breathing method is abdominal breathing or chest breathing is closely related to the sleep depth of the subject, and thus the accuracy of estimating the sleep depth can be improved.
- the sleep state estimation device of the present invention it is possible to more easily classify the subject's body movement in more detail, and to improve the accuracy of estimating the state of the subject such as sleep depth and body movement. This can be realized more easily.
- the sleep state estimation device of this embodiment is a device that is mounted on a vehicle or the like, estimates a sleep state such as a sleep depth or body motion of a subject, and performs countermeasures using various application programs.
- the sleep state estimation device 1 of the present embodiment includes a seat 10, a respiration sensor 12, an acceleration sensor 14, an I / F 20, an arithmetic unit 30, and a DB 40.
- the respiration sensor 12 is a non-invasive sensor that measures the respiration of the subject M.
- the respiration sensor 12 includes a plurality of pressure sensors installed on the seating surface and the backrest of the seat 10.
- the respiration sensor 12 has a pressure sensor 13 a that measures movement of the chest of the subject M on the seat 10, a pressure sensor 13 b that measures movement of the abdomen of the subject M, and a seating surface of the subject M.
- the pressure sensor 13c which measures the pressure of is provided.
- the sheet 10 may include respiration bands 16 and 18 for measuring the respiration of the subject M.
- the breathing bands 16 and 18 are strain gauge bands that can detect the breathing of the subject M.
- the breathing band 16 detects the movement of the subject M's rib cage.
- the breathing band 18 detects the movement of the abdominal cavity of the subject M.
- the seat 10 may include a heart rate sensor and an electroencephalogram sensor separately.
- the seat 10 includes an acceleration sensor 14 that detects the acceleration of the vehicle.
- the acceleration sensor 14 a sensor mounted in advance in the vehicle can be used.
- the acceleration sensor 14 can use one prepared in the sleep state estimation device 1.
- the detection values of the respiration sensor 12 and the acceleration sensor 14 are processed by the arithmetic unit 30 via the I / F 20.
- the arithmetic unit 30 while referring to the DB 40 in which data for each individual subject M is recorded, the sleep depth of the subject M, deep breathing, conversation, re-sitting of the subject M, stretching both hands upward, left and right
- the sleep state of the target person M is estimated as to whether or not body movement such as lifting the buttocks is performed.
- the arithmetic unit 30 of the sleep state estimation device 1 determines whether a feature value (described later) of a respiratory waveform by the respiratory sensor 12 or the like is normal or abnormal (S01).
- the arithmetic unit 30 determines that the abnormality of the feature value of the respiration waveform is based on the detected values of the respiration sensor 12 and the acceleration sensor 14 such as deep breathing of the subject M. It is determined whether it is caused by body movement or caused by vehicle behavior such as vehicle braking (S02).
- the arithmetic unit 30 When the abnormality of the feature value of the respiratory waveform is caused by the vehicle behavior (S02), the arithmetic unit 30 indicates that the abnormality of the feature value of the respiratory waveform is caused by the vehicle behavior, and the subject M's It is determined that it is impossible to estimate body movement (S03). In these operations of S01 to S03, filter processing is performed with the output waveforms of the respiration sensor 12 and the acceleration sensor.
- the arithmetic unit 30 determines whether the subject M is deeply sleep or in a state other than deep sleep (S04). When it is determined that the subject M is deeply asleep, the arithmetic unit 30 outputs that fact (S04, S05). When the feature quantity of the respiratory waveform is abnormal (S01), and the abnormality of the feature quantity of the respiratory waveform is caused by the body movement of the subject M (S02), or the feature quantity of the respiratory waveform is normal (S01). ) When the state of the subject M is a state other than deep sleep (S04), the arithmetic unit 30 determines whether the subject M is awake or light sleep (S06).
- the arithmetic unit 30 outputs a determination result of whether the subject M is awake or sleeps (S06 to S08). In these operations of S04 to S08, the arithmetic unit 30 refers to the learning data stored in the DB 40 or estimates the sleep state of the subject M while storing new learning data in the DB 40.
- the arithmetic unit 30 can determine whether the subject M is awake or sleeping. In this case, when it is determined that the subject person M is sleeping (S04), the arithmetic unit outputs a computation result indicating that the subject person M is sleeping in the above S05.
- the arithmetic unit When the feature quantity of the respiratory waveform is normal (S01) and the subject M is awake (S04), or the feature quantity of the respiratory waveform is abnormal (S01), the abnormality of the feature quantity of the respiratory waveform is the target When it is caused by the body movement of the person M (S02), or the arithmetic unit 30 outputs a calculation result indicating that the subject person M is awake.
- the arithmetic unit 30 recognizes the subject M and sets reproducibility, amplitude, period, and autocorrelation thresholds a to d, which are feature quantities of a respiratory waveform described later (S101). These threshold values a to d are set for each subject M and are recorded in the DB 40.
- the arithmetic unit 30 uses the detection value of the acceleration sensor 14 to determine whether or not the X component of acceleration (the longitudinal direction of the vehicle) exceeds a threshold value such as 1.0 m / s 2 (S102).
- the arithmetic unit 30 uses the detection value of the acceleration sensor 14 to determine whether the Y component of the acceleration (the left-right direction of the vehicle) exceeds a threshold value such as 1.0 m / s 2 (S103).
- the arithmetic unit 30 estimates the detection value by the respiration sensor 12 as vehicle noise ( S104).
- the arithmetic unit 30 calculates the feature amount from the detection value by the respiration sensor 12. Reproducibility is calculated (S105).
- Reproducibility is the fluctuation of the minimum value for each cycle of the respiratory waveform. If the respiration waveform as shown in FIG. 6 is detected by the respiration sensor 12, the minimum values for each period of the respiration waveform are ⁇ , ⁇ , and ⁇ .
- the arithmetic unit 30 determines whether or not the reproducibility F exceeds the threshold value a (S106). When the reproducibility F exceeds the threshold value a (S106), the arithmetic unit 30 estimates that the subject M is awake and identifies the body motion of the subject M as described later (S113).
- the arithmetic unit 30 calculates the amplitude as the feature amount from the detection value by the respiration sensor 12 (S107). As shown in FIG. 7, the amplitude of the respiratory waveform is obtained from the difference between the lowest value and the highest value of the respiratory waveform in the immediately preceding cycle.
- the arithmetic unit 30 determines whether or not the amplitude exceeds the threshold value b (S108). When the amplitude exceeds the threshold value b (S108), the arithmetic unit 30 estimates that the subject M is awake and identifies the body movement of the subject M as described later (S113).
- the arithmetic unit 30 calculates a period as a feature amount from the detection value by the respiratory sensor 12 (S109). The arithmetic unit 30 determines whether or not the cycle exceeds the threshold value c (S110). When the period exceeds the threshold value c (S110), the arithmetic unit 30 estimates that the subject M is awake and identifies the body movement of the subject M as will be described later (S113).
- the arithmetic unit 30 calculates autocorrelation as a feature amount (S111).
- Autocorrelation is the identity of a waveform obtained by shifting the respiratory waveform by an arbitrary time and the original respiratory waveform. For example, as shown in FIG. 8, the respiratory waveform detected by the respiratory sensor 12 is shifted by 2 seconds, 5 seconds, or 1 to 3 cycles, and the difference between the smooth waveform and the original respiratory waveform is digitized. The autocorrelation is calculated.
- the difference between the waveform obtained by shifting the respiratory waveform by a certain period and the original respiratory waveform, and the difference between the waveform obtained by shifting the respiratory waveform by a different period and the original respiratory waveform are quantified.
- the accuracy of detection can be improved by assuming that there is a change in autocorrelation.
- the arithmetic unit 30 determines whether or not the autocorrelation exceeds the threshold value d (S112). When the autocorrelation exceeds the threshold value d (S112), the arithmetic unit 30 estimates that the subject M is awake and identifies the body movement of the subject M as described later (S113). When the autocorrelation is equal to or less than the threshold value d (S112), the arithmetic unit 30 estimates that the subject M is in a sleep state and identifies the sleep stage and body movement of the subject M as will be described later (S114). ).
- the present inventor obtained knowledge that the body movement of the subject M can be estimated with a considerably high probability by using the combination of the amplitude, period, autocorrelation and reproducibility of the respiratory waveform of the subject M. It was. For example, as shown in FIG. 9, it can be seen that there are considerable differences in the amplitude, reproducibility, etc. of the respiration waveform of the subject M depending on the body movement of the subject M such as deep breathing and vehicle behavior such as braking. .
- the body movement of the subject M is estimated by using a combination of the amplitude, period, autocorrelation and reproducibility of the breathing waveform of the subject M.
- FIG. 10 when the subject M sits down, changes appear in the amplitude, autocorrelation and reproducibility of the respiratory waveform.
- the subject M extends his hand upward, changes appear in the amplitude and reproducibility of the respiratory waveform.
- the subject M has a conversation, a change appears in the autocorrelation and reproducibility of the respiratory waveform.
- the subject M takes a deep breath, a change appears in the cycle and amplitude of the breathing waveform.
- the present inventor also makes use of a combination of features such as the amplitude of the respiratory waveform and autocorrelation for body movements in which the subject M rearranges or stretches the legs. The knowledge that it is possible to detect with high accuracy was obtained.
- Each waveform is the raw waveform, period, amplitude, autocorrelation between the waveform before 2 periods and the current respiration waveform, and autocorrelation between the waveform before 3 periods and the current respiration waveform and reproducibility. Respectively.
- the numerical values calculated for easy comparison are normalized. When the numerical value is 0, there is no change from normal, and when the numerical value exceeds 0, there is a change from normal. It shows that. When the normalized numerical value exceeds 1, 1 is set as the upper limit value. As shown in FIG. 11, there is a significant difference in the period and amplitude of the respiratory waveform during deep breathing indicated by the broken line, compared to the normal time indicated by the solid line.
- the accuracy of detecting the state of the subject M is improved by removing noise using the detected value by the acceleration sensor 14 in addition to the respiratory waveform of the subject M and its feature amount.
- the calculation unit 30 as an estimator gave an erroneous estimation result if the sleep depth of the target person M is shallow. Even in this case, if the body movement is estimated from the respiratory waveform, the estimation result can be corrected because the possibility that the subject M is awake is high.
- the arithmetic unit 30 as an estimator may generate a vehicle based on the respiratory waveform even if an erroneous estimation result is obtained if the sleep depth of the subject person M is deep. If the behavior is estimated or the vehicle behavior is detected by the acceleration sensor 14, it can be determined that the reliability of the estimation result that the sleep depth is estimated based on the respiratory waveform is low, and the reason for the erroneous estimation result can be verified. Available. If the body movement or vehicle behavior of the subject M cannot be detected by the respiratory waveform or the acceleration sensor 14, the erroneous estimation result may not be corrected, but the body motion or vehicle behavior of the subject M may not be corrected. By always detecting, it is possible to improve the estimation accuracy by correcting the estimation result.
- a primary filter based on the detection positions of the plurality of respiration sensors 12 provided on the seat 10 is provided to roughly classify whether the sleep state is deep or shallow.
- abdominal respiration increases in a deep sleep state
- a chest respiration increases in a shallow sleep state. Therefore, in this embodiment, it is determined whether the breathing method of the subject M is abdominal breathing or chest breathing from the detection positions of the plurality of breathing sensors 12 provided on the seat 10, and the sleep state of the subject M is determined.
- the classification is deep or shallow with high accuracy. Further, since the human body movement tends to be stable (rest) in the deep sleep stage, it is possible to estimate the sleep depth based on the dispersion of the measurement positions.
- ⁇ Breathing methods can be broadly divided into chest breathing and abdominal breathing, and the elements of both are often mixed unconsciously.
- the mental state of the subject M is strongly related to breathing, and breathing is shallower and shorter when the mind and body are in tension. In other words, when there is a sense of tension or anxiety, humans tend to become unconsciously chest breathing.
- the abdominal breathing becomes dominant in deep sleep because the whole body is relaxed and relaxed while sleeping.
- the amount of air entering and exiting at a time is several times different between chest-type breathing and abdominal-type breathing, which is 500 ml for chest-type breathing and 2000 ml at maximum for abdominal-type breathing.
- the respiration which is one of the observable physiological indices, has a lower frequency when the sleep becomes deeper and the frequency becomes slower than when the sleep is active or shallow.
- the pressure sensors 13 a and 13 b of the respiration sensor 12 are arranged at least on the upper and lower stages of the seat 10 so that the chest and abdomen of the subject M can be measured. Are classified into deep sleep and other states according to changes in the respiratory type of abdominal breathing a and chest breathing b.
- breathing bands 16 and 18 as shown in FIG. 3 may be used.
- the arithmetic unit 30 of the sleep state estimation device 1 calculates the amplitude difference or phase difference between the abdominal breathing and the chest breathing once or every breath, takes an average value for 30 seconds, and takes 30 seconds. The coefficient of variation is obtained in the processing interval, the magnitude relationship between the breast type and the abdominal type is compared, and differential integration is performed in the processing interval to calculate a vector.
- FIG. 18 shows the subject's abdominal breathing a, chest breathing b, and sleep stage d. As shown in FIG. 18, it can be seen that the abdominal breathing a and the chest breathing b are antagonized whenever the sleep stage d changes.
- a section as shown in FIG. 20 is defined. Sections every 30 seconds are defined as sections P1 to P3 in order. In addition, a section extending over the sections P1 and P2 is defined as a section P4, and a section extending over the sections P2 and P3 is defined as a section P5.
- the change in the numerical value of the sleep stage is merely an example and has no relation to the actual data. Further, observed values of abdominal breathing a, chest breathing b, and sleep stage d as shown in FIG. 21 are obtained.
- both abdominal breathing a and chest breathing b tend to fall as shown in FIG. 22.
- chest breathing b tends to drop and abdominal breathing tends to rise.
- abdominal breathing a and chest breathing b tend to rise.
- abdominal breathing b tends to rise and the abdominal breathing tends to fall.
- FIG. 23 shows the change in the ratio of the abdominal breathing a and the chest breathing b for each of the sections P1 to P5 defined in FIG. 20, and the numerical values are shown in the table of FIG.
- the ratio of the interval P1 to the interval P2 of the chest breathing is indicated by b (P1 / P2)
- the ratio of the interval P2 to the interval P3 of the abdominal breathing is indicated by a (P2 / P3).
- FIG. 25 shows the ratio of the section P1 to the section P2 of abdominal breathing for each change in sleep stage.
- the sleep stage can be estimated from simple changes in abdominal respiration and chest respiration.
- a change in the breathing method for example, the ratio of abdominal breathing to the chest breathing (abdominal / chest type), deep sleep in the sleep stage and other state changes can be captured.
- the variance (standard deviation) of the ratio of the abdominal breathing to the chest breathing (abdomen / chest) it is possible to improve the separation accuracy of the awake state and the sleep state in the sleep stage.
- the arithmetic unit 30 of the sleep state estimation apparatus 1 estimates the state of the subject based on the identity of the subject M for each period of the respiratory waveform as the feature quantity of the respiratory waveform. Since the subject M himself / herself can arbitrarily control the respiratory fluctuation, the body of the subject M is estimated by estimating the state of the subject M based on the identity of the subject M for each period of the respiratory waveform. It becomes possible to classify the movement in more detail more easily, and it becomes easier to improve the accuracy of estimating the state of the subject M such as sleep depth and body movement.
- the arithmetic unit 30 of the sleep state estimation device 1 uses the waveform obtained by shifting the respiratory waveform by an arbitrary time and the original respiratory waveform as the identity of the respiratory waveform of the subject M.
- the state of the subject M is estimated based on at least one of the autocorrelation that is the same as the above and the reproducibility that is the fluctuation of the minimum value for each cycle of the respiratory waveform.
- the arithmetic unit 30 of the sleep state estimation apparatus 1 is based on the period, amplitude, autocorrelation, and reproducibility of the respiration waveform as the characteristic amount of the respiration waveform of the revelation person M. Estimate the state. As a result, the state of the subject M is estimated by combining the four indicators of the respiratory waveform period, amplitude, autocorrelation and reproducibility, so the state of the subject M such as sleep depth and body motion is estimated. It is possible to further improve the accuracy of performing.
- the arithmetic unit 30 of the sleep state estimation apparatus 1 detects that the amplitude, autocorrelation, and reproducibility of the respiratory waveform are fluctuating compared to the normal time of the subject M. If so, it is estimated that the subject M is in a sitting position.
- the present inventor has obtained knowledge that when the subject M is in a sitting state, fluctuations in the amplitude, autocorrelation and reproducibility of the breathing waveform of the subject M are observed. For this reason, when it is detected that the amplitude, autocorrelation and reproducibility of the respiration waveform are fluctuating as compared with the normal time of the subject M, it is estimated that the subject M is sitting again. By doing so, it is possible to accurately estimate the state in which the subject M is sitting back.
- the arithmetic unit 30 of the sleep state estimation device 1 detects that the amplitude and reproducibility of the respiration waveform are fluctuating as compared with the normal time of the subject, the subject M moves his hand upward. Presumed to be stretched.
- the present inventor has obtained knowledge that when the subject M is in a state of reaching his hand upward, fluctuations in the amplitude and reproducibility of the breathing waveform of the subject M are observed. For this reason, when it is detected that the amplitude and reproducibility of the respiration waveform are fluctuating as compared with the normal time of the subject M, it is estimated that the subject M is in a state of reaching out upward. Thus, it is possible to accurately estimate the state in which the subject M is reaching up.
- the arithmetic unit 30 of the sleep state estimation device 1 detects that the autocorrelation and reproducibility of the respiratory waveform are fluctuating as compared with the normal time of the subject M, the subject M is talking. It is presumed that the user is in a state.
- the present inventor has obtained the knowledge that when the subject M is in a conversation state, the autocorrelation and reproducibility of the breathing waveform of the subject M are changed. For this reason, when it is detected that the autocorrelation and reproducibility of the respiration waveform are fluctuating compared to the normal time of the subject M, it is estimated that the subject M is in a conversation state. Thus, it is possible to accurately estimate the state in which the subject M is talking.
- the arithmetic unit 30 of the sleep state estimation device 1 detects that the period and amplitude of the respiratory waveform are fluctuating compared to the normal time of the subject M, the target It is estimated that the person M is taking a deep breath.
- the present inventor has obtained knowledge that when the subject M is taking a deep breath, fluctuations are observed in the period and amplitude of the breathing waveform of the subject M. For this reason, when it detects that the period and amplitude of the respiration waveform M are fluctuating compared with the normal time of the subject M, it is estimated that the subject M is taking a deep breath. The state in which the subject M is taking a deep breath can be accurately estimated.
- the arithmetic unit 30 of the sleep state estimation apparatus 1 estimates the state of the subject M by comparing the feature amount of the respiratory waveform and the threshold value set for each feature amount, By the processing, it becomes possible to estimate the state of the subject M such as sleep depth and body movement.
- the arithmetic unit 30 of the sleep state estimation device 1 sets a threshold value for each feature amount for each subject M. For this reason, it becomes possible to estimate the state of the subject M corresponding to the constitution and preference for each subject M.
- the arithmetic unit 30 of the sleep state estimation apparatus 1 estimates the state of the subject person M who boarded the vehicle, and based on the acceleration of the vehicle, the vehicle behavior of the vehicle and the body movement of the subject person M are calculated.
- the state of the subject M is estimated while discriminating. For this reason, it becomes possible to accurately estimate the state of the subject M who has boarded the vehicle while discriminating the vehicle behavior and the body movement.
- the arithmetic unit 30 of the sleep state estimation apparatus 1 determines whether the breathing method of the subject M is abdominal breathing or chest breathing from the breathing waveform of the subject M, The sleep depth of the subject M is estimated. Whether the breathing method is the abdominal breathing or the chest breathing is closely related to the sleep depth of the subject M, and thus the accuracy of estimating the sleep depth can be improved.
- the sleep state estimation device of the present invention it is possible to more easily classify the subject's body movement in more detail, and to improve the accuracy of estimating the state of the subject such as sleep depth and body movement. This can be realized more easily. For this reason, it is easy to execute various application programs on the subject in accordance with the estimated depth of sleep of the subject and the type of body motion, etc., and lead the subject to a more comfortable state. It becomes.
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Veterinary Medicine (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Physics & Mathematics (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Public Health (AREA)
- Physiology (AREA)
- Psychiatry (AREA)
- Anesthesiology (AREA)
- Pulmonology (AREA)
- Developmental Disabilities (AREA)
- Educational Technology (AREA)
- Hospice & Palliative Care (AREA)
- Child & Adolescent Psychology (AREA)
- Psychology (AREA)
- Social Psychology (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
Description
10 シート
12 呼吸センサ
13a,13b,13c 圧力センサ
14 加速度センサ
16,18 呼吸バンド
20 I/F
30 演算ユニット
40 DB
Claims (11)
- 対象者の呼吸波形の特徴量として前記呼吸波形の周期ごとの同一性に基づいて前記対象者の状態を推定する状態推定ユニットを備えた睡眠状態推定装置。
- 前記状態推定ユニットは、前記対象者の前記呼吸波形の周期ごとの同一性として、前記呼吸波形を任意の時間だけずらした波形と元の前記呼吸波形との同一性である自己相関性と、前記呼吸波形の周期ごとの最低値のゆらぎである再現性との少なくともいずれかに基づいて前記対象者の状態を推定する、請求項1に記載の睡眠状態推定装置。
- 前記状態推定ユニットは、前記対象者の前記呼吸波形の前記特徴量として前記呼吸波形の周期、振幅、前記自己相関性及び前記再現性に基づいて前記対象者の状態を推定する、請求項2に記載の睡眠状態推定装置。
- 前記状態推定ユニットは、前記呼吸波形の振幅、前記自己相関性及び前記再現性が、前記対象者の通常時と比較して変動していることを検出した場合は、前記対象者が座り直している状態であると推定する、請求項3に記載の睡眠状態推定装置。
- 前記状態推定ユニットは、前記呼吸波形の振幅及び前記再現性が、前記対象者の通常時と比較して変動していることを検出した場合は、前記対象者が上方に手を伸ばしている状態であると推定する、請求項3又は4に記載の睡眠状態推定装置。
- 前記状態推定ユニットは、前記呼吸波形の前記自己相関性及び前記再現性が、前記対象者の通常時と比較して変動していることを検出した場合は、前記対象者が会話をしている状態であると推定する、請求項3~5のいずれか1項に記載の睡眠状態推定装置。
- 前記状態推定ユニットは、前記呼吸波形の周期及び振幅が、前記対象者の通常時と比較して変動していることを検出した場合は、前記対象者が深呼吸をしている状態であると推定する、請求項3~6のいずれか1項に記載の睡眠状態推定装置。
- 前記状態推定ユニットは、前記呼吸波形の前記特徴量と前記特徴量ごとに設定された閾値とを比較することで前記対象者の状態を推定する、請求項1~7のいずれか1項に記載の睡眠状態推定装置。
- 前記状態推定ユニットは、前記対象者ごとに前記特徴量ごとの前記閾値を設定する、請求項8に記載の睡眠状態推定装置。
- 前記状態推定ユニットは、車両に搭乗した前記対象者の状態を推定し、前記車両の加速度に基づいて前記車両の車両挙動と前記対象者の体動とを判別しつつ前記対象者の状態を推定する、請求項1~9のいずれか1項に記載の睡眠状態推定装置。
- 前記状態推定ユニットは、前記対象者の前記呼吸波形から前記対象者の呼吸方式が腹式呼吸であるか胸式呼吸であるかを判定することにより、前記対象者の睡眠深度を推定する、請求項1~10のいずれか1項に記載の睡眠状態推定装置。
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2010/069847 WO2012063308A1 (ja) | 2010-11-08 | 2010-11-08 | 睡眠状態推定装置 |
JP2012542734A JP5454703B2 (ja) | 2010-11-08 | 2010-11-08 | 睡眠状態推定装置 |
EP10859585.1A EP2638855A4 (en) | 2010-11-08 | 2010-11-08 | DEVICE FOR SLEEP STATE ESTIMATION |
US13/884,118 US20130231579A1 (en) | 2010-11-08 | 2010-11-08 | Sleep state estimation device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2010/069847 WO2012063308A1 (ja) | 2010-11-08 | 2010-11-08 | 睡眠状態推定装置 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2012063308A1 true WO2012063308A1 (ja) | 2012-05-18 |
Family
ID=46050491
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2010/069847 WO2012063308A1 (ja) | 2010-11-08 | 2010-11-08 | 睡眠状態推定装置 |
Country Status (4)
Country | Link |
---|---|
US (1) | US20130231579A1 (ja) |
EP (1) | EP2638855A4 (ja) |
JP (1) | JP5454703B2 (ja) |
WO (1) | WO2012063308A1 (ja) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105658144A (zh) * | 2013-10-21 | 2016-06-08 | 提爱思科技股份有限公司 | 清醒装置、座椅及清醒度判断方法 |
JP2020000371A (ja) * | 2018-06-26 | 2020-01-09 | 凸版印刷株式会社 | 睡眠状態判定装置、睡眠状態判定方法、及び睡眠状態判定システム |
JP2020519381A (ja) * | 2017-05-08 | 2020-07-02 | マドナニ, アカーシュMADNANI, Akash | 人間のパフォーマンスを観察するシステムおよび方法 |
WO2024140869A1 (zh) * | 2022-12-29 | 2024-07-04 | 华为技术有限公司 | 数据处理方法、装置、设备及计算机可读存储介质 |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP6124011B2 (ja) * | 2013-10-21 | 2017-05-10 | テイ・エス テック株式会社 | 覚醒装置及びシート |
US10449874B2 (en) * | 2015-03-27 | 2019-10-22 | Ts Tech Co., Ltd. | Seat with detector |
CN106308752A (zh) * | 2016-08-23 | 2017-01-11 | 广东小天才科技有限公司 | 一种基于可穿戴设备的睡眠监测方法和系统 |
CN106361282A (zh) * | 2016-08-31 | 2017-02-01 | 珠海多士科技有限公司 | 睡眠质量检测方法与系统 |
EP4349250A3 (en) | 2017-12-22 | 2024-06-26 | ResMed Sensor Technologies Limited | Apparatus, system, and method for motion sensing |
KR102658390B1 (ko) | 2017-12-22 | 2024-04-17 | 레스메드 센서 테크놀로지스 리미티드 | 건강 및 의료 감지를 위한 장치, 시스템, 및 방법 |
EP3727145B1 (en) * | 2017-12-22 | 2024-01-24 | ResMed Sensor Technologies Limited | Apparatus, system, and method for physiological sensing in vehicles |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2005312653A (ja) * | 2004-04-28 | 2005-11-10 | Denso Corp | 運転者状態検出装置及びプログラム |
JP2006055501A (ja) * | 2004-08-23 | 2006-03-02 | Denso Corp | 眠気検出装置及び方法 |
JP2007175225A (ja) * | 2005-12-27 | 2007-07-12 | Sumitomo Osaka Cement Co Ltd | 状態解析装置及びソフトウエアプログラム |
JP2007236488A (ja) * | 2006-03-06 | 2007-09-20 | Toyota Motor Corp | 覚醒度推定装置及びシステム並びに方法 |
JP2010088725A (ja) | 2008-10-09 | 2010-04-22 | Daikin Ind Ltd | 睡眠判定装置 |
JP2010220649A (ja) * | 2009-03-19 | 2010-10-07 | Toyota Motor Corp | 睡眠装置及び睡眠維持方法 |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070179395A1 (en) * | 2002-08-07 | 2007-08-02 | Sotos John G | System and method for assessment of sleep |
JP3627741B2 (ja) * | 2003-03-06 | 2005-03-09 | 松下電器産業株式会社 | 睡眠時呼吸情報測定装置 |
JP5210676B2 (ja) * | 2008-03-19 | 2013-06-12 | パナソニックヘルスケア株式会社 | 温度制御装置、空調機および電気毛布 |
JPWO2009150744A1 (ja) * | 2008-06-13 | 2011-11-10 | ハートメトリクス株式会社 | 睡眠状態モニタリング装置、モニタリングシステムおよびコンピュータプログラム |
JP2010122732A (ja) * | 2008-11-17 | 2010-06-03 | Aisin Seiki Co Ltd | 安全運転支援システム |
-
2010
- 2010-11-08 JP JP2012542734A patent/JP5454703B2/ja not_active Expired - Fee Related
- 2010-11-08 EP EP10859585.1A patent/EP2638855A4/en not_active Withdrawn
- 2010-11-08 WO PCT/JP2010/069847 patent/WO2012063308A1/ja active Application Filing
- 2010-11-08 US US13/884,118 patent/US20130231579A1/en not_active Abandoned
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2005312653A (ja) * | 2004-04-28 | 2005-11-10 | Denso Corp | 運転者状態検出装置及びプログラム |
JP2006055501A (ja) * | 2004-08-23 | 2006-03-02 | Denso Corp | 眠気検出装置及び方法 |
JP2007175225A (ja) * | 2005-12-27 | 2007-07-12 | Sumitomo Osaka Cement Co Ltd | 状態解析装置及びソフトウエアプログラム |
JP2007236488A (ja) * | 2006-03-06 | 2007-09-20 | Toyota Motor Corp | 覚醒度推定装置及びシステム並びに方法 |
JP2010088725A (ja) | 2008-10-09 | 2010-04-22 | Daikin Ind Ltd | 睡眠判定装置 |
JP2010220649A (ja) * | 2009-03-19 | 2010-10-07 | Toyota Motor Corp | 睡眠装置及び睡眠維持方法 |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105658144A (zh) * | 2013-10-21 | 2016-06-08 | 提爱思科技股份有限公司 | 清醒装置、座椅及清醒度判断方法 |
CN105658144B (zh) * | 2013-10-21 | 2019-06-21 | 提爱思科技股份有限公司 | 清醒装置、座椅及清醒度判断方法 |
JP2020519381A (ja) * | 2017-05-08 | 2020-07-02 | マドナニ, アカーシュMADNANI, Akash | 人間のパフォーマンスを観察するシステムおよび方法 |
JP2024010157A (ja) * | 2017-05-08 | 2024-01-23 | マドナニ,アカーシュ | ユーザの健康状態を評価するためのシステムおよび前記システムの作動方法 |
JP2020000371A (ja) * | 2018-06-26 | 2020-01-09 | 凸版印刷株式会社 | 睡眠状態判定装置、睡眠状態判定方法、及び睡眠状態判定システム |
WO2024140869A1 (zh) * | 2022-12-29 | 2024-07-04 | 华为技术有限公司 | 数据处理方法、装置、设备及计算机可读存储介质 |
Also Published As
Publication number | Publication date |
---|---|
JP5454703B2 (ja) | 2014-03-26 |
EP2638855A1 (en) | 2013-09-18 |
US20130231579A1 (en) | 2013-09-05 |
EP2638855A4 (en) | 2015-09-30 |
JPWO2012063308A1 (ja) | 2014-05-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5454703B2 (ja) | 睡眠状態推定装置 | |
JP6594399B2 (ja) | 生体情報モニタリングシステム | |
JP3733133B2 (ja) | 睡眠状態推定装置 | |
JP3976752B2 (ja) | 睡眠状態推定装置及びプログラム | |
JP4993980B2 (ja) | 呼気時間を出力可能な装置および方法 | |
WO2010143535A1 (ja) | 眠気判定装置 | |
CN108888271A (zh) | 一种生理参数测量系统及配备该测量系统的智能座椅 | |
JP5352814B2 (ja) | 自律神経成分指標推定装置及び自律神経成分指標推定方法 | |
WO2005082252A1 (ja) | 睡眠段階判定方法 | |
WO2019111977A1 (ja) | 姿勢判定装置 | |
WO2013125048A1 (ja) | 睡眠品質推定装置、睡眠品質推定方法及び睡眠品質推定用プログラム | |
JP2009297474A (ja) | 睡眠段階判定装置 | |
JPWO2004107978A1 (ja) | 睡眠段階判定方法および判定装置 | |
WO2016076253A1 (ja) | 睡眠状態判定装置、睡眠状態判定方法及びプログラム | |
JP6589108B2 (ja) | 無呼吸及び/又は低呼吸診断装置 | |
KR101853102B1 (ko) | 가속도 센서 기반 수면분류 정보 측정기 | |
JP2011160852A (ja) | 覚醒状態検出装置 | |
JP2009028239A (ja) | 居眠り運転検出装置 | |
JP2008080071A (ja) | 睡眠の質評価装置 | |
JP5652764B2 (ja) | 睡眠状態判定装置及び睡眠状態判定プログラム | |
JP6738458B2 (ja) | 睡眠状態判定装置及び睡眠状態判定方法 | |
JP2014183994A (ja) | 体動判定装置 | |
JP6518294B2 (ja) | 睡眠評価装置及びプログラム | |
JP6011240B2 (ja) | 生体情報取得方法及び生体情報取得装置 | |
JP7591717B2 (ja) | いきみ検出装置 |
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: 10859585 Country of ref document: EP Kind code of ref document: A1 |
|
DPE1 | Request for preliminary examination filed after expiration of 19th month from priority date (pct application filed from 20040101) | ||
ENP | Entry into the national phase |
Ref document number: 2012542734 Country of ref document: JP Kind code of ref document: A |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 13884118 Country of ref document: US |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2010859585 Country of ref document: EP |