CN116440425B - Intelligent adjusting method and system of LED photodynamic therapeutic instrument - Google Patents
Intelligent adjusting method and system of LED photodynamic therapeutic instrument Download PDFInfo
- Publication number
- CN116440425B CN116440425B CN202310725849.4A CN202310725849A CN116440425B CN 116440425 B CN116440425 B CN 116440425B CN 202310725849 A CN202310725849 A CN 202310725849A CN 116440425 B CN116440425 B CN 116440425B
- Authority
- CN
- China
- Prior art keywords
- information
- target
- generate
- data
- wavelength
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000001225 therapeutic effect Effects 0.000 title claims abstract description 242
- 238000000034 method Methods 0.000 title claims abstract description 90
- 230000035790 physiological processes and functions Effects 0.000 claims abstract description 155
- 230000001815 facial effect Effects 0.000 claims abstract description 137
- 238000013507 mapping Methods 0.000 claims abstract description 103
- 238000012549 training Methods 0.000 claims abstract description 74
- 238000012545 processing Methods 0.000 claims abstract description 67
- 238000013506 data mapping Methods 0.000 claims abstract description 12
- 238000002428 photodynamic therapy Methods 0.000 claims description 59
- 230000008569 process Effects 0.000 claims description 43
- 230000014509 gene expression Effects 0.000 claims description 22
- 238000004458 analytical method Methods 0.000 claims description 19
- 230000006870 function Effects 0.000 claims description 19
- 230000009466 transformation Effects 0.000 claims description 16
- 238000000605 extraction Methods 0.000 claims description 14
- 238000004364 calculation method Methods 0.000 claims description 13
- 238000010276 construction Methods 0.000 claims description 11
- 238000007621 cluster analysis Methods 0.000 claims description 10
- 230000010354 integration Effects 0.000 claims description 5
- 238000002560 therapeutic procedure Methods 0.000 claims description 2
- 238000013473 artificial intelligence Methods 0.000 abstract description 2
- 238000004422 calculation algorithm Methods 0.000 description 20
- 230000008921 facial expression Effects 0.000 description 16
- 230000036772 blood pressure Effects 0.000 description 10
- 239000000284 extract Substances 0.000 description 10
- 238000010801 machine learning Methods 0.000 description 9
- 230000000694 effects Effects 0.000 description 8
- 230000003287 optical effect Effects 0.000 description 8
- 238000011282 treatment Methods 0.000 description 8
- 238000001126 phototherapy Methods 0.000 description 6
- 230000003190 augmentative effect Effects 0.000 description 5
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 4
- 239000008280 blood Substances 0.000 description 4
- 210000004369 blood Anatomy 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000001914 filtration Methods 0.000 description 4
- 239000001301 oxygen Substances 0.000 description 4
- 229910052760 oxygen Inorganic materials 0.000 description 4
- 230000001105 regulatory effect Effects 0.000 description 4
- 230000037393 skin firmness Effects 0.000 description 4
- 238000007619 statistical method Methods 0.000 description 4
- 206010061218 Inflammation Diseases 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 3
- 239000006185 dispersion Substances 0.000 description 3
- 230000002996 emotional effect Effects 0.000 description 3
- 230000004054 inflammatory process Effects 0.000 description 3
- 238000011369 optimal treatment Methods 0.000 description 3
- 206010002368 Anger Diseases 0.000 description 2
- 206010020772 Hypertension Diseases 0.000 description 2
- 238000003066 decision tree Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 239000007789 gas Substances 0.000 description 2
- 238000003064 k means clustering Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000004962 physiological condition Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000036555 skin type Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000037303 wrinkles Effects 0.000 description 2
- 208000019901 Anxiety disease Diseases 0.000 description 1
- WQZGKKKJIJFFOK-GASJEMHNSA-N Glucose Natural products OC[C@H]1OC(O)[C@H](O)[C@@H](O)[C@@H]1O WQZGKKKJIJFFOK-GASJEMHNSA-N 0.000 description 1
- 230000036506 anxiety Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000004071 biological effect Effects 0.000 description 1
- 230000036760 body temperature Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013075 data extraction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 230000008451 emotion Effects 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 210000001097 facial muscle Anatomy 0.000 description 1
- 239000008103 glucose Substances 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000011337 individualized treatment Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000035515 penetration Effects 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000036387 respiratory rate Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000037394 skin elasticity Effects 0.000 description 1
- 230000003860 sleep quality Effects 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 238000011426 transformation method Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- 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
- A61N5/0613—Apparatus adapted for a specific treatment
- A61N5/062—Photodynamic therapy, i.e. excitation of an agent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/174—Facial expression recognition
-
- 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
- G16H20/30—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to physical therapies or activities, e.g. physiotherapy, acupressure or exercising
-
- 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
- G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
- G16H40/60—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices
-
- 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
- A61N2005/0626—Monitoring, verifying, controlling systems and methods
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
- Y02B20/00—Energy efficient lighting technologies, e.g. halogen lamps or gas discharge lamps
- Y02B20/40—Control techniques providing energy savings, e.g. smart controller or presence detection
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Primary Health Care (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Biophysics (AREA)
- General Engineering & Computer Science (AREA)
- Epidemiology (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Animal Behavior & Ethology (AREA)
- Veterinary Medicine (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Mathematical Physics (AREA)
- Pathology (AREA)
- Radiology & Medical Imaging (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Physical Education & Sports Medicine (AREA)
- General Business, Economics & Management (AREA)
- Business, Economics & Management (AREA)
- Radiation-Therapy Devices (AREA)
Abstract
The invention relates to the field of artificial intelligence, and discloses an intelligent adjusting method and system of an LED photodynamic therapeutic apparatus, which are used for improving the accuracy and efficiency of intelligent adjustment of the LED photodynamic therapeutic apparatus. Comprising the following steps: generating a plurality of therapeutic instrument parameter dimensions, collecting physiological state information and facial information, constructing a mapping table, and generating a target mapping table; performing data expansion on the face information to generate expanded face information, performing data mapping to generate expanded physiological state information, performing standardization processing to generate a training data set; model training and parameter adjustment are carried out on the initial therapeutic instrument intelligent adjustment model, and a target therapeutic instrument intelligent adjustment model is obtained; collecting current facial information to predict the wavelength of the therapeutic instrument and generating a target wavelength; and performing power matching on the LED photodynamic therapeutic instrument to generate target power, and adjusting the LED photodynamic therapeutic instrument based on the target wavelength and the target power.
Description
Technical Field
The invention relates to the field of artificial intelligence, in particular to an intelligent adjusting method and system of an LED photodynamic therapeutic apparatus.
Background
Currently, the LED photodynamic therapeutic apparatus is widely applied to a plurality of medical fields such as ophthalmology, dermatology, stomatology and the like, has the characteristics of noninvasive, high efficiency, reliability and the like, and is favored by a plurality of medical staff and patients.
However, for the problem of wavelength prediction in LED photodynamic therapy devices, most of the prior art methods are limited relatively only by starting from physiological state information of the patient; secondly, the existing intelligent regulation technology is based on rules, and lacks adaptability and accuracy to real scenes; in addition, there are some problems of data processing and feature extraction, such as individual differences of data, selection of features, lack of training data sets, and the like. Therefore, the accuracy in intelligent adjustment and wavelength prediction of the LED photodynamic therapy device is low at present.
Disclosure of Invention
The invention provides an intelligent adjusting method and system of an LED photodynamic therapeutic apparatus, which are used for improving the accuracy and efficiency of intelligent adjustment of the LED photodynamic therapeutic apparatus.
The first aspect of the invention provides an intelligent adjustment method of an LED photodynamic therapeutic apparatus, which is used for collecting working parameter information of the LED photodynamic therapeutic apparatus in a working process, extracting parameter dimensions of the working parameter information and generating a plurality of therapeutic apparatus parameter dimensions;
Collecting physiological state information of a target user, and collecting facial information of the target user at the same time;
building a mapping table of the physiological state information and the facial information to generate a target mapping table;
performing data expansion on the face information to generate expanded face information;
performing data mapping on the extended face information based on the target mapping table to generate extended physiological state information;
carrying out standardization processing on the extended physiological state information and the working parameter information to generate a training data set;
model training is carried out on a preset intelligent regulation model of the initial therapeutic apparatus through the training data set, and parameter regulation is carried out on the intelligent regulation model of the initial therapeutic apparatus through a preset loss function, so that an intelligent regulation model of the target therapeutic apparatus is obtained;
collecting current facial information of the target user, inputting the current facial information into the intelligent regulation model of the target therapeutic instrument to predict the wavelength of the therapeutic instrument, and generating a target wavelength;
and performing power matching on the LED photodynamic therapeutic instrument based on the target wavelength to generate target power, and adjusting the LED photodynamic therapeutic instrument based on the target wavelength and the target power.
With reference to the first aspect, in a first implementation manner of the first aspect of the present invention, the collecting working parameter information of the LED photodynamic therapeutic apparatus in a working process, extracting a parameter dimension of the working parameter information, and generating a plurality of therapeutic apparatus parameter dimensions includes:
collecting power data of the LED photodynamic therapeutic instrument to generate power data;
collecting wavelength data of the LED photodynamic therapeutic instrument to generate wavelength data;
collecting working time of the LED photodynamic therapeutic instrument to generate time data;
performing parameter integration on the power data, the wavelength data and the time data to generate working parameter data;
traversing the working parameter data parameter dimension to generate a plurality of therapeutic instrument parameter dimensions.
With reference to the first implementation manner of the first aspect, in a second implementation manner of the first aspect of the present invention, the mapping table construction is performed on the physiological status information and the facial information, and a target mapping table is generated, which includes:
performing cluster analysis on the physiological state information to generate a corresponding cluster feature set;
extracting heart rate data of the target user through the cluster feature set to generate target heart rate data;
Extracting the facial information by using the user expression to generate corresponding user expression information;
carrying out mapping relation analysis on the user expression data through the target heart rate data to generate a corresponding target mapping relation;
and constructing a mapping table of the physiological state information and the facial information through the target mapping relation to generate a target mapping table.
With reference to the first aspect, in a third implementation manner of the first aspect of the present invention, the performing data expansion on the face information to generate expanded face information includes:
performing noise processing on the face information to generate candidate face information;
performing data scaling processing on the candidate face information to generate scaled face information;
and carrying out mirror symmetry processing on the scaled face information to generate extended face information.
With reference to the first aspect, in a fourth implementation manner of the first aspect of the present invention, the normalizing the extended physiological state information and the operating parameter information to generate a training data set includes:
carrying out data combination on the extended physiological state information and the working parameter information to generate a data set to be processed;
Performing decentration on the data set to be processed to generate a zero-axis distribution data set;
performing standard deviation calculation on the zero-axis distribution data set to generate a corresponding standard deviation;
scaling the zero-axis distribution data set by the standard deviation to generate a scaled data set;
and carrying out logarithmic transformation processing on the scaled data set to generate a corresponding training data set.
With reference to the first aspect, in a fifth implementation manner of the first aspect of the present invention, the collecting current face information of the target user, and inputting the current face information to the target therapeutic apparatus intelligent regulation model to perform therapeutic apparatus wavelength prediction, generating a target wavelength includes:
collecting current facial information of the target user, extracting facial feature points of the current facial information through the intelligent regulation model of the target therapeutic instrument, and generating a plurality of facial feature points;
performing physiological state information mapping on the current facial information through the facial feature points to generate corresponding current physiological state information;
and predicting the wavelength of the therapeutic instrument according to the current physiological state information to generate a target wavelength.
With reference to the fifth implementation manner of the first aspect, in a sixth implementation manner of the first aspect of the present invention, the performing therapeutic apparatus wavelength prediction according to the current physiological state information, generating a target wavelength includes:
Carrying out therapeutic instrument power data prediction on the LED photodynamic therapeutic instrument through the current physiological state information to generate predicted power data;
predicting the working time length of the LED photodynamic therapeutic instrument according to the predicted power data to generate predicted working time length;
and calculating the wavelength of the therapeutic instrument based on the predicted power data and the predicted working time length, and generating a target wavelength.
The second aspect of the present invention provides an intelligent regulation system of an LED photodynamic therapy apparatus, the intelligent regulation system of the LED photodynamic therapy apparatus comprising:
the extraction module is used for collecting working parameter information of the LED photodynamic therapeutic instrument in the working process, extracting parameter dimensions of the working parameter information and generating a plurality of therapeutic instrument parameter dimensions;
the acquisition module is used for acquiring physiological state information of a target user and acquiring facial information of the target user at the same time;
the construction module is used for constructing a mapping table of the physiological state information and the facial information and generating a target mapping table;
the expansion module is used for carrying out data expansion on the face information to generate expanded face information;
the mapping module is used for carrying out data mapping on the extended face information based on the target mapping table to generate extended physiological state information;
The generating module is used for carrying out standardized processing on the extended physiological state information and the working parameter information to generate a training data set;
the training module is used for carrying out model training on a preset intelligent regulation model of the initial therapeutic apparatus through the training data set, and carrying out parameter regulation on the intelligent regulation model of the initial therapeutic apparatus through a preset loss function to obtain an intelligent regulation model of the target therapeutic apparatus;
the input module is used for acquiring the current facial information of the target user, inputting the current facial information into the intelligent regulation model of the target therapeutic instrument to predict the wavelength of the therapeutic instrument and generating a target wavelength;
and the matching module is used for carrying out power matching on the LED photodynamic therapeutic instrument based on the target wavelength, generating target power and adjusting the LED photodynamic therapeutic instrument based on the target wavelength and the target power.
A third aspect of the present invention provides an intelligent regulation apparatus for an LED photodynamic therapy device, comprising: a memory and at least one processor, the memory having instructions stored therein; the at least one processor invokes the instructions in the memory to cause the intelligent regulation device of the LED photodynamic therapy apparatus to execute the intelligent regulation method of the LED photodynamic therapy apparatus described above.
A fourth aspect of the present application provides a computer readable storage medium having instructions stored therein that, when run on a computer, cause the computer to perform the intelligent adjustment method of an LED photodynamic therapy device described above.
According to the technical scheme provided by the application, working parameter information of the LED photodynamic therapeutic apparatus in the working process is collected, parameter dimension extraction is carried out on the working parameter information, a plurality of therapeutic apparatus parameter dimensions are generated, physiological state information of a target user is collected, and meanwhile, face information of the target user is collected; building a mapping table of the physiological state information and the facial information to generate a target mapping table; performing data expansion on the face information to generate expanded face information; performing data mapping on the extended face information based on the target mapping table to generate extended physiological state information; carrying out standardized processing on the extended physiological state information and the working parameter information to generate a training data set; model training is carried out on a preset intelligent regulation model of the initial therapeutic apparatus through a training data set, and parameter regulation is carried out on the intelligent regulation model of the initial therapeutic apparatus through a preset loss function, so that an intelligent regulation model of the target therapeutic apparatus is obtained; collecting current face information of a target user, inputting the current face information into an intelligent regulation model of a target therapeutic instrument to predict the wavelength of the therapeutic instrument, and generating a target wavelength; and performing power matching on the LED photodynamic therapeutic instrument based on the target wavelength to generate target power, and adjusting the LED photodynamic therapeutic instrument based on the target wavelength and the target power. According to the application, the optimal treatment wavelength and power parameters are obtained by collecting physiological state information, facial information and working parameter information of the LED photodynamic therapeutic apparatus of a target user and predicting the LED wavelength based on the prediction model, and a series of parameters such as facial information, physiological state and the like are fused together by establishing the intelligent regulation model of the target therapeutic apparatus to form an intelligent therapeutic apparatus regulating system, so that the intelligent degree of the therapeutic apparatus is improved. By carrying out data expansion and mapping table construction of physiological state information on the facial information, a more complete and comprehensive training data set can be obtained, and the prediction accuracy and robustness of the model are improved. The physiological state information and the working parameter information are subjected to standardized processing, so that the data among different characteristics can be coordinated and consistent, and the processibility of the data and the stability of the model are improved. The LED photodynamic therapeutic instrument is adjusted based on the target wavelength and the power, the self-adaptive adjustment can be performed according to facial features, physiological states and photodynamic therapeutic requirements of different patients, and the accuracy and the efficiency of intelligent adjustment of the LED photodynamic therapeutic instrument are improved.
Drawings
FIG. 1 is a schematic diagram of an embodiment of a method for intelligent adjustment of an LED photodynamic therapy device according to an embodiment of the present invention;
FIG. 2 is a flowchart of mapping table construction for physiological status information and facial information according to an embodiment of the present invention;
FIG. 3 is a flowchart of data expansion of face information according to an embodiment of the present invention;
FIG. 4 is a flowchart of the normalization of the extended physiological status information and the operating parameter information according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of an embodiment of an intelligent regulation system of an LED photodynamic therapy device according to an embodiment of the present invention;
fig. 6 is a schematic diagram of an embodiment of an intelligent adjusting apparatus of an LED photodynamic therapy device according to an embodiment of the present invention.
Detailed Description
The embodiment of the invention provides an intelligent adjusting method and an intelligent adjusting system for an LED photodynamic therapeutic apparatus, which are used for improving the accuracy and the efficiency of intelligent adjustment of the LED photodynamic therapeutic apparatus. The terms "first," "second," "third," "fourth" and the like in the description and in the claims and in the above drawings, if any, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments described herein may be implemented in other sequences than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or apparatus.
For ease of understanding, a specific flow of an embodiment of the present invention is described below with reference to fig. 1, and an embodiment of a method for intelligently adjusting an LED photodynamic therapy device according to the embodiment of the present invention includes:
s101, collecting working parameter information of an LED photodynamic therapeutic apparatus in a working process, extracting parameter dimensions of the working parameter information, and generating a plurality of therapeutic apparatus parameter dimensions;
it is to be understood that the execution body of the present invention may be an intelligent adjustment system of an LED photodynamic therapy apparatus, and may also be a terminal or a server, which is not limited herein. The embodiment of the invention is described by taking a server as an execution main body as an example.
Specifically, in the process of acquiring the working parameter information of the LED photodynamic therapeutic apparatus, power data, wavelength data and working time data can be acquired, and parameter integration is performed on the data to generate working parameter data. The server collects power data of the LED photodynamic therapy device, which can be achieved through a sensor or a measuring device. The power data is indicative of the power level output by the therapeutic device during operation. The server then collects wavelength data, which may be acquired by a spectrometer or sensor. The wavelength data indicates the wavelength of light used by the therapeutic apparatus. The server collects the working time data, and the working time of the therapeutic instrument can be recorded through a timer or a recorder. The collected power data, wavelength data, and time data may be integrated into an operating parameter data set. For example, assume that the server has collected 10 power data points, 1W, 2W, 1.5W, 2.5W, 1.8W, 1.2W, 2.2W, 1.6W, 1.9W, and 2.3W, respectively; 10 wavelength data points were collected, 450nm, 470nm, 460nm, 480nm, 465nm, 455nm, 475nm, 462nm, 467nm, and 472nm, respectively; 10 time data points were acquired, 10 seconds, 20 seconds, 15 seconds, 25 seconds, 18 seconds, 12 seconds, 22 seconds, 16 seconds, 19 seconds, and 23 seconds, respectively. Integrating these data results in an operating parameter data set. Then, the server performs parameter dimension traversal on the working parameter data set to generate a plurality of therapeutic instrument parameter dimensions. The parameter dimension is a mode of classifying and dividing the working parameters, and can be defined according to actual requirements. For example, the server divides data into three dimensions, low power, medium power, and high power, according to power level; dividing the data into three dimensions of blue light, green light and red light according to the wavelength; data are divided into short, medium and long dimensions according to time. Thus, the server obtains a plurality of therapeutic instrument parameter dimensions. For example, assume that the server divides the power data in three dimensions of low power (1W, 1.5W, 1.8W, 1.2W, 1.6W, and 1.9W), medium power (2W, 2.2W, and 2.3W), and high power (2.5W); wavelength data were divided in three dimensions, blue (450 nm, 460nm, 455nm and 462 nm), green (470 nm, 465nm and 467 nm) and red (480 nm, 475nm and 472 nm); the time data is divided into three dimensions of short time (10 seconds, 15 seconds, and 12 seconds), medium time (20 seconds, 18 seconds, and 16 seconds), and long time (25 seconds, 22 seconds, 19 seconds, and 23 seconds). Through such parameter dimension traversal, the server obtains multiple therapeutic instrument parameter dimensions, such as low power + blue + short, medium power + green + medium, high power + red + long, etc. The parameter dimension extraction and combination can provide richer data information for the subsequent intelligent adjustment method, so that the LED photodynamic therapeutic instrument can be adjusted according to different parameter dimensions, and personalized therapeutic effects are realized. The server can better understand and control the working state of the LED photodynamic therapeutic instrument by collecting the working parameter information and generating a plurality of therapeutic instrument parameter dimensions, thereby laying a foundation for realizing an intelligent regulation method.
S102, collecting physiological state information of a target user, and collecting facial information of the target user;
specifically, in order to collect physiological state information and facial information of the target user, the server uses different sensors, devices and technical means simultaneously. Such an integrated data collection method can provide comprehensive and accurate user information. For example, the server uses devices such as a heart rate monitor and a blood pressure monitor to monitor heart rate and blood pressure data of the target user in real time. Meanwhile, the server measures the blood oxygen saturation of the target user through a pulse oxygen saturation meter or a handheld oximeter. In addition, body temperature data of the target user can be acquired through the thermometer. In acquiring face information, a server acquires a face image of a target user using a camera or a depth sensor or the like. Through facial recognition techniques, the server extracts facial features such as skin conditions, facial expressions, wrinkles, and the like. Further, biosensors and wearable devices such as heart rate monitors, motion trackers, sleep monitors, etc. can also be used to obtain more physiological state information such as amount of exercise, sleep quality, respiratory rate, etc. By simultaneously collecting physiological state information and facial information, the server obtains comprehensive target user data, thereby better understanding the physical condition and facial features thereof. The comprehensive data acquisition method is beneficial to research and application in the fields of personalized medicine, facial analysis, health monitoring and the like.
S103, constructing a mapping table of the physiological state information and the facial information, and generating a target mapping table;
specifically, the physiological state information is subjected to cluster analysis, similar data points are divided into the same category, and a corresponding cluster feature set is generated. The cluster analysis may use a machine learning algorithm, such as K-means clustering or hierarchical clustering, to categorize the data according to its similarity. For example, assume that the server collects heart rate and blood pressure data of the target user. Through cluster analysis, the server divides the data points into different clusters, such as high heart rate-hypertension, low heart rate-hypertension, normal heart rate-normal blood pressure, and the like. Next, the server extracts heart rate data of the target user by clustering the feature set. For each cluster, the server calculates an average or other statistical indicator of its heart rate data as a representation of the target heart rate data. And simultaneously, extracting the user expression from the facial information. Through facial expression recognition technology, the server analyzes the facial image of the target user, and extracts expression information such as smiling face, angry face or surprise face. And then, analyzing through the mapping relation between the target heart rate data and the user expression data to generate a corresponding target mapping relation. For example, the server finds that at high heart rates, the target user more likely exhibits an expression of gas or anxiety, while at normal heart rates, the target user more likely exhibits a pleasant or relaxed expression. Based on the target mapping relation, the server builds a mapping table, integrates the physiological state information and the facial information, and generates a target mapping table. This mapping table may correspond different physiological state information and facial information to specific target features or behaviors. For example, assume that through cluster analysis, the server divides the heart rate data of the target user into two clusters of high heart rate and normal heart rate. Then, through facial expression extraction, the server obtains two expression data of the smiling face and the angry face of the user. By analyzing this data, the server finds that a face of qi is more likely to appear at a high heart rate, and a smiling face is more likely to appear at a normal heart rate. Based on this mapping relationship, the server builds a target mapping table, maps the high heart rate and the face of the raw gas to a "tension state", and maps the normal heart rate and the face of the smile to a "relaxation state". In this way, the server generates a target mapping table, which predicts the state of the target user, such as tension or relaxation, based on the physiological state information and facial information of the target user. Through the above steps, the server successfully builds the mapping table and generates the target mapping table, correlating the physiological status information and the facial information. The mapping table provides a basis for a subsequent intelligent adjustment method, so that the LED photodynamic therapy instrument can be intelligently adjusted according to the physiological state and the facial expression of a target user.
S104, carrying out data expansion on the face information to generate expanded face information;
specifically, the face information is subjected to noise processing, and candidate face information is generated. The noise processing can adopt filtering or noise reduction algorithm to remove noise interference in the face image, so as to obtain clear candidate face information. For example, mean filtering, median filtering, or gaussian filtering methods may be used to smooth the facial image and reduce noise. Next, data scaling processing is performed on the candidate face information, generating scaled face information. The data scaling may change the size of the face image by adjusting its size. This may scale the candidate face images to different sizes using image processing techniques, such as interpolation algorithms. By scaling the face information, extended face information having different scales and levels of detail can be generated. Then, mirror symmetry processing is performed on the scaled face information, and extended face information is generated. The mirror symmetry processing refers to horizontally turning over the face image to obtain a mirror image thereof. This may be done symmetrically by an image processing algorithm, such as a flipping operation or a mirror transformation. Through the mirror symmetry processing, extended face information having different sides and symmetry can be generated. For example, assume that the server has a candidate face image, and clear face information is obtained after noise processing. The server then performs a scaling process on the candidate face information to generate a smaller face image. And then, the server performs mirror symmetry processing on the scaled face image to obtain a mirror image of the face image. In this way, the server generates augmented face information including the original face image, the scaled face image, and the mirrored face image. By carrying out data expansion on the facial information, the server increases the diversity and the number of data samples, and improves the generalization capability and the robustness of the model. This helps training the intelligent regulation model better, makes the LED photodynamic therapy appearance can adapt to different facial features and demand, provides more individualized treatment effect.
S105, performing data mapping on the extended face information based on the target mapping table to generate extended physiological state information;
specifically, based on the target mapping table, the server performs data mapping on the extended face information to generate extended physiological state information. In this process, the server first prepares a target mapping table containing mapping relationships between different physiological state information and facial information. This mapping table may be predefined or learned from training data by a machine learning algorithm. Next, the server has extended face information including original face information, extended face information obtained after noise processing, data scaling, and mirror symmetry processing. The goal of the server is to map these augmented facial information to corresponding physiological state information. To achieve this, the server uses facial expression recognition techniques to extract expression information in the augmented facial information, such as smiles, angers, surprise, and the like. Then, the server maps the extracted expression information into corresponding physiological state information by using the mapping relation between the facial information and the physiological state information in the target mapping surface. For example, assume that the target mapping table of the server maps smiles to a relaxed state. When the server extracts the expression of the extended face information, if smiles are extracted, the server maps the expression information to a relaxed state according to the definition of the target mapping table. Thus, the server obtains corresponding physiological state information from the extended face information. Through the data mapping based on the target mapping table, the server can correlate the extended facial information with the physiological state information, and more accurate and personalized guidance is provided for the subsequent intelligent regulation method. The intelligent adjusting method can automatically adjust the parameters of the LED photodynamic therapeutic apparatus according to the real-time facial information of the user so as to realize more effective and personalized therapeutic effects.
S106, carrying out standardized processing on the extended physiological state information and the working parameter information to generate a training data set;
specifically, the extended physiological state information and the working parameter information are subjected to data combination to generate a data set to be processed. This set of data to be processed contains a combination of extended physiological state information and operating parameter information. For example, the server represents the extended physiological state information as a vector containing parameters such as heart rate, blood pressure, etc., while the operating parameter information may contain parameters such as power, wavelength, time, etc. Next, a decentralization process is performed on the data set to be processed in order to shift the center of the data set to the origin position. Decentralization may be achieved by calculating the average of the data sets and subtracting the average from each data point. This eliminates the overall offset of the data set, making the data distribution more symmetrical with respect to the origin. Then, standard deviation calculation is performed on the zero-axis distribution data set to obtain the standard deviation of the data set. The standard deviation is an index that measures the degree of discretization of the data set, reflecting the average deviation between the data point and the mean. By calculating the standard deviation, the server knows the distribution of the data and the degree of difference between the data points. The zero-axis distribution dataset is then scaled using the standard deviation to generate a scaled dataset. Scaling is the linear transformation of data points to bring their range to a particular scale. By dividing each data point of the data set by the standard deviation, the data point can be scaled to a suitable range for subsequent processing and analysis. To further enhance the processibility of the data, a logarithmic transformation process may be performed on the scaled data set. The logarithmic transformation is to take the logarithm of the value of the data point to reduce the amplitude difference of the data so as to make the data more in line with the normal distribution. This may make the data more comparable and interpretable, facilitating training of the model or further statistical analysis. In summary, by the standardized processing of the extended physiological state information and the operating parameter information, the server generates a training dataset having uniform dimensions and distribution, providing a reliable basis for subsequent model training and analysis. For example, assume that the server has augmented physiological state information including heart rate and blood pressure, and operating parameter information including power and wavelength. The server combines the information and performs the de-centering, standard deviation calculation, scaling and logarithmic transformation processes to finally obtain a standardized data set for training, wherein each sample has the same data range and distribution characteristics. For example, assume that the server has a target user, and that the server has collected his augmented physiological state information, including a heart rate of 75 beats/minute and a blood pressure of 120/80 mmHg, and operating parameter information, including a power of 10 watts and a wavelength of 650 nanometers. These information are combined to form the data set to be processed. Next, the server performs a decentralization process on the data set to be processed. An average of the extended physiological state information and the operating parameter information is calculated and subtracted from each data point. The average heart rate was assumed to be 70 times/min, the average blood pressure was 110/70 mmHg, the average power was 8 Watts, and the average wavelength was 600 nanometers. The center of the dataset is moved to the origin position by the de-centering process. Then, it is served more in conformity with the normal distribution. For example, the server takes the logarithm of the heart rate, blood pressure, power, and wavelength values. Through the above processing, the server obtains the training data set subjected to the normalization processing. Each sample in the dataset is subjected to merging, decentralization, standard deviation calculation, scaling and logarithmic transformation, and has uniform scale and distribution characteristics. Such training data sets may be used to train an intelligent adjustment model to enable the LED photodynamic therapy device to be individually adjusted according to the physiological state and operating parameters of the user.
S107, performing model training on a preset intelligent regulation model of the initial therapeutic apparatus through a training data set, and performing parameter regulation on the intelligent regulation model of the initial therapeutic apparatus through a preset loss function to obtain an intelligent regulation model of the target therapeutic apparatus;
specifically, the server uses the training data set as input data, which includes standardized extended physiological state information and operating parameter information. Each sample in these datasets is associated with a desired target wavelength and power. Next, the server selects the appropriate model architecture to train the smart tuning model. This may be a neural network, decision tree, support vector machine, etc. machine learning algorithm. According to specific requirements and data characteristics, a proper model architecture is selected. The server then uses the training data set to model the initial therapeutic intelligent regulation model. During the training process, the model learns patterns and rules in the data set, and the relationship between the physiological state information and the working parameter information and the target wavelength. By iterative optimization algorithms (e.g., gradient descent), the parameters of the model are continually adjusted to minimize the preset loss function. The loss function is a measure of the difference between the model predicted result and the actual target value. The server defines a suitable loss function to evaluate the performance of the model. For example, the mean square error (Mean Squared Error) can be used as a loss function to calculate the average of the squared differences between the model predicted and target values. During training, the model adjusts the parameters according to the feedback signal of the loss function to minimize the value of the loss function. In this way, the model gradually optimizes and improves the predictive power of the target wavelength. For example, assume that the server has a training data set that includes extended physiological state information and operating parameter information, and a target wavelength corresponding thereto. The server selects a neural network model as the initial therapeutic intelligent regulation model and defines the mean square error as the loss function. By inputting the training data set into the neural network model, the model will learn based on the characteristics and patterns of the data. During the training process, the model will continually adjust the parameters to minimize the value of the loss function. Through repeated iterative training, the model can be gradually optimized, and has better wavelength prediction capability. For example, in the initial stage, the model may have a large error in the prediction of the target wavelength. However, as training proceeds, the model gradually adjusts the parameters by observing the correlation between the physiological state information and the operating parameter information in the training data and the target wavelength to reduce the prediction error. The model can more accurately predict the target wavelength through the learned mode and rule. For example, the model may find that a higher power value is better suited to achieve a desired target wavelength under a particular physiological condition. By adjusting the parameters of the model, the power of the therapeutic apparatus can be intelligently adjusted according to physiological state information and working parameter information so as to achieve the effect of being closer to the target wavelength. During the training process, the server uses a portion of the training data as a validation set for evaluating the performance of the model. And by monitoring the loss function value on the verification set, the server judges whether the model is over-fitted or under-fitted and adjusts the model according to the requirement. After repeated iterative training, parameters of the model can converge to an optimal value, so that the intelligent regulation model of the target therapeutic instrument is obtained. The model has the capability of intelligently predicting the wavelength according to the physiological state information and the working parameter information of the target user and correspondingly adjusting the wavelength.
S108, collecting current face information of a target user, inputting the current face information into an intelligent regulation model of a target therapeutic instrument to predict the wavelength of the therapeutic instrument, and generating a target wavelength;
specifically, the server collects current face information of the target user. This may be achieved by using a camera or other facial recognition device to acquire facial images or video of the user. The facial information may include facial features, expressions, skin conditions, and the like. Next, the server processes the current facial information using the target therapeutic apparatus intelligent regulation model. The server identifies and extracts a plurality of key facial feature points from the facial image using a facial feature point extraction algorithm in the model. These feature points may represent important features of the face, such as eyes, nose, mouth, etc. Then, the server maps the current face information to corresponding physiological state information through the facial feature points. This mapping process may be performed according to a previously constructed target mapping table. For example, the location and shape of particular facial feature points may be associated with certain physiological states, such as skin firmness, degree of wrinkles, and the like. By analyzing the position and morphology of these feature points, the server deduces the current physiological state information of the user. And through the current physiological state information, the server predicts the wavelength of the therapeutic instrument by using the intelligent regulation model of the target therapeutic instrument. The model predicts the wavelength of the therapeutic instrument which is most suitable for the target user through calculation and analysis according to the input of the current physiological state information. This predicted wavelength can be optimized based on previous training and models to maximize the fit to the physiological state and needs of the user. For example, assume that a server is developing an LED photodynamic therapy device for skin care. The camera is used for acquiring facial images of target users, and the server uses a facial feature point extraction algorithm in the intelligent regulation model of the target therapeutic instrument to extract key feature points such as eyes, nose, mouth and the like from the images. From the previously constructed target mapping table, the server knows that the location and shape of the particular ocular feature point is related to skin firmness. The server deduces the skin compactness physiological state information of the target user by analyzing the positions and the forms of the eye feature points of the target user. The physiological state information is input into an intelligent regulation model of the target therapeutic instrument, and the model predicts the wavelength of the therapeutic instrument which is most suitable for the target user according to the knowledge and the mode obtained by training. For example, the model may predict that higher wavelength phototherapy is needed to promote skin firmness and elasticity based on skin firmness physiological state information of the user. Thus, the server obtains the prediction result of the target wavelength by collecting the current facial information of the target user and inputting the current facial information into the intelligent regulation model of the target therapeutic instrument for processing and analysis. This prediction may guide the setting of the therapeutic apparatus to provide the best suited phototherapy effect for the user. Meanwhile, the collection of training data needs to fully consider the diversity and representativeness of the data to obtain reliable model prediction results. In summary, by collecting current facial information of a target user and inputting the current facial information to a target therapeutic apparatus intelligent regulation model, the wavelength prediction of the therapeutic apparatus can be realized. The process involves facial feature point extraction, physiological state information mapping, model prediction and other steps, and can provide accurate treatment parameter setting for personalized LED photodynamic treatment. The power of the LED photodynamic therapeutic apparatus is matched based on the target wavelength, and the LED photodynamic therapeutic apparatus is regulated according to the target wavelength and the target power so as to ensure that the light parameters used in treatment are matched with the required treatment effect. And predicting therapeutic instrument power data of the LED photodynamic therapeutic instrument according to the current physiological state information. The server maps the current physiological state information to corresponding predicted power data according to a previously constructed target mapping table. This mapping may be established by training a data set and a machine learning algorithm. For example, based on physiological state information (e.g., skin type, degree of inflammation, etc.), the model may predict an appropriate power level to maximize therapeutic effect. Next, the LED photodynamic therapy device is subjected to operation duration prediction by predicting power data. According to the specification and performance characteristics of the therapeutic apparatus, a relation model can be established between the predicted power data and the working time. This model may take into account factors such as the power consumption rate, battery capacity, etc. of the therapeutic device to predict the expected duration that the therapeutic device may operate at a given power level. Based on the predicted power data and the predicted operating time, a therapeutic instrument wavelength may be calculated to generate a target wavelength. This calculation may be performed according to previously set algorithms and rules. For example, based on the predicted power data and the predicted operating time, the most appropriate wavelength setting may be selected in combination with the characteristics of the therapeutic device and the light source wavelength range to provide the best therapeutic effect. For example, assume that a server is developing an LED photodynamic therapy device for the treatment of inflammation. The server maps the current physiological state information of the target user to the predicted power data through a previously established target mapping table. For example, by analyzing information about the type of skin, the degree of inflammation and the sensitivity of the user, the model can predict an appropriate power level, such as 50mW. Next, the server predicts the working time by predicting the power data according to the specifications and performance characteristics of the therapeutic apparatus. The battery capacity of the therapeutic apparatus was assumed to be 2000mAh and the power consumption rate was assumed to be 10mW/min. And according to the predicted power data of 50mW, the server calculates that the predicted working time of the therapeutic apparatus is 200 minutes. Based on the predicted power data of 50mW and the predicted operating time of 200 minutes, in combination with the wavelength range of the therapeutic apparatus (e.g., 400-700 nm), the most appropriate wavelength setting may be selected to provide the best therapeutic effect. In this example, the server determines, based on previous research and clinical experimental results, that the phototherapy effect using a wavelength of 450nm is optimal at a power of 50mW. Therefore, the server obtains the prediction result of the target wavelength by performing therapeutic instrument power data prediction, working time length prediction and wavelength calculation on the current physiological state information. This target wavelength will guide the setting of the LED photodynamic therapy device, ensuring that the most appropriate phototherapy effect is provided in a given physiological state. In summary, the wavelength of the therapeutic apparatus can be predicted by collecting the current facial information of the target user and inputting the current facial information into the intelligent regulation model of the target therapeutic apparatus for processing and analysis. The process involves facial feature point extraction, physiological state information mapping, power data prediction, working time length prediction, wavelength calculation, and the like, to provide personalized LED photodynamic therapy settings in order to achieve optimal therapeutic effects.
And S109, performing power matching on the LED photodynamic therapeutic apparatus based on the target wavelength, generating target power, and adjusting the LED photodynamic therapeutic apparatus based on the target wavelength and the target power.
Specifically, by the aforementioned method, the server has obtained the predicted result of the target wavelength. Next, the server determines a target power for the LED photodynamic therapy device based on the target wavelength. The target power refers to the output power required by the therapeutic device at a given wavelength. In order to achieve power matching, the server performs experiments and tests in advance to determine the power effect relationship at different wavelengths. This can be done by testing a series of wavelengths to measure the therapeutic effect at different powers. According to the experimental result, the server establishes a correlation model between wavelength and power. For example. Assuming that the server determines through experiments that the optimal therapeutic effect can be obtained when the phototherapy power is 60mW at the wavelength of 450 nm. Then, in case the predicted target wavelength is 450nm, the server sets the target power to 60mW to achieve power matching. Once the target power is determined, the server may adjust the LED photodynamic therapy device according to the target wavelength and the target power. This can be achieved by controlling the output optical power of the therapeutic apparatus. The therapeutic device may have different control means, such as adjusting current, adjusting voltage or adjusting light intensity, etc. By adjusting the output light power of the therapeutic apparatus to achieve the target power, it is ensured that the LED photodynamic therapeutic apparatus used provides the desired therapeutic effect at a given wavelength. In this way, the server can adjust the settings of the therapeutic device based on the predicted target wavelength and target power to maximize the delivery of the customized therapy.
In the embodiment of the application, working parameter information of the LED photodynamic therapeutic apparatus in the working process is collected, parameter dimension extraction is carried out on the working parameter information, a plurality of therapeutic apparatus parameter dimensions are generated, physiological state information of a target user is collected, and meanwhile, facial information of the target user is collected; building a mapping table of the physiological state information and the facial information to generate a target mapping table; performing data expansion on the face information to generate expanded face information; performing data mapping on the extended face information based on the target mapping table to generate extended physiological state information; carrying out standardized processing on the extended physiological state information and the working parameter information to generate a training data set; model training is carried out on a preset intelligent regulation model of the initial therapeutic apparatus through a training data set, and parameter regulation is carried out on the intelligent regulation model of the initial therapeutic apparatus through a preset loss function, so that an intelligent regulation model of the target therapeutic apparatus is obtained; collecting current face information of a target user, inputting the current face information into an intelligent regulation model of a target therapeutic instrument to predict the wavelength of the therapeutic instrument, and generating a target wavelength; and performing power matching on the LED photodynamic therapeutic instrument based on the target wavelength to generate target power, and adjusting the LED photodynamic therapeutic instrument based on the target wavelength and the target power. According to the application, the optimal treatment wavelength and power parameters are obtained by collecting physiological state information, facial information and working parameter information of the LED photodynamic therapeutic apparatus of a target user and predicting the LED wavelength based on the prediction model, and a series of parameters such as facial information, physiological state and the like are fused together by establishing the intelligent regulation model of the target therapeutic apparatus to form an intelligent therapeutic apparatus regulating system, so that the intelligent degree of the therapeutic apparatus is improved. By carrying out data expansion and mapping table construction of physiological state information on the facial information, a more complete and comprehensive training data set can be obtained, and the prediction accuracy and robustness of the model are improved. The physiological state information and the working parameter information are subjected to standardized processing, so that the data among different characteristics can be coordinated and consistent, and the processibility of the data and the stability of the model are improved. The LED photodynamic therapeutic instrument is adjusted based on the target wavelength and the power, the self-adaptive adjustment can be performed according to facial features, physiological states and photodynamic therapeutic requirements of different patients, and the accuracy and the efficiency of intelligent adjustment of the LED photodynamic therapeutic instrument are improved.
In a specific embodiment, the process of executing step S101 may specifically include the following steps:
(1) Collecting power data of the LED photodynamic therapeutic instrument to generate power data;
(2) Collecting wavelength data of the LED photodynamic therapeutic instrument to generate wavelength data;
(3) Collecting working time of the LED photodynamic therapeutic instrument to generate time data;
(4) Carrying out parameter integration on the power data, the wavelength data and the time data to generate working parameter data;
(5) Traversing the working parameter data parameter dimension to generate a plurality of therapeutic instrument parameter dimensions.
Specifically, a proper sensor configuration and data acquisition system for the therapeutic apparatus is first required. These sensors may be used to measure optical power and therapeutic effects, such as optical power meters, photosensitive detectors, laser spectrometers, and the like. Through these sensors, the server monitors the optical power and the success rate in real time. For example, assume that the server uses one optical power meter to measure the output power of the LED photodynamic therapy device. The server connects the optical power meter with the therapeutic apparatus and sets a proper sampling frequency for data acquisition. During each treatment, the optical power meter will record the value of the output power. These power data can be used for subsequent analysis and processing. Similarly, to acquire wavelength data, a laser spectrometer or other wavelength measurement device may be used. These devices can measure the wavelength characteristics of the light output by the LED photodynamic therapy device. By collecting the wavelength data, the server knows the wavelength range and the wavelength distribution used by the therapeutic apparatus. In addition, for the acquisition of the working time, a timer or recording of the working state of the therapeutic apparatus may be used for acquisition. The server calculates the length of time for each treatment by recording the on and off times of the therapeutic apparatus. These time data are very useful for subsequent analysis and statistics. Once the power data, wavelength data, and time data are obtained, the server integrates them into operating parameter data. This can be achieved by integrating and combining these data in different parameter dimensions of the therapeutic apparatus. For example, the server combines the power data, wavelength data, and time data into one dataset containing these parameters. Thus, each data sample will contain a power value, a wavelength value, and a time value that represent the operating state of the therapeutic device at a particular point in time. For example, assume that the server has a set of power data, including [50mW, 60mW, 55mW ], a set of wavelength data, including [450nm, 470nm, 460nm ], and a set of time data, including [30min, 40min, 35min ]. The server integrates them into one working parameter data set: power data: [50mW, 60mW, 55mW ]; wavelength data: [450nm, 470nm, 460nm ]; time data: [30min, 40min, 35min ]. The integrated operating parameter data set is as follows: [50mW, 450nm, 30min ]; [60mW, 470nm, 40min ]; [55mW,460nm, 35min ]. By means of parameter integration, the server obtains an operating parameter data set comprising power, wavelength and time. Each sample represents a different combination of therapeutic parameters. In this way, the server can use this data for subsequent analysis, modeling, and optimization.
In a specific embodiment, as shown in fig. 2, the step of performing S103 may specifically include the following steps:
s201, carrying out cluster analysis on physiological state information to generate a corresponding cluster feature set;
s202, extracting heart rate data of a target user through a cluster feature set to generate target heart rate data;
s203, extracting the user expression of the facial information to generate corresponding user expression information;
s204, carrying out mapping relation analysis on the user expression data through target heart rate data to generate a corresponding target mapping relation;
s205, mapping table construction is carried out on the physiological state information and the facial information through the target mapping relation, and a target mapping table is generated.
It should be noted that, the physiological status information is subjected to cluster analysis, and physiological status data points with similar characteristics are categorized into the same cluster. The clustering algorithm can be selected according to specific situations, such as K-means, hierarchical clustering and the like. The result of the cluster analysis will generate corresponding cluster feature sets, each feature set representing a cluster. Next, heart rate data extraction is performed on the target user by the cluster feature set. For each clustered feature set, heart rate data corresponding to the feature set is extracted from the physiological state information of the target user. This may be achieved by calculating the average, peak or spectral analysis of the heart rate of the target user over a particular period of time. And simultaneously, extracting the user expression from the facial information. Facial information of the target user is analyzed and processed using computer vision techniques, such as facial expression recognition algorithms, and facial expression information, such as smiles, angers, surprise, etc., is extracted. This can be achieved by detecting the position of facial feature points, analyzing the degree of activity of facial muscles, and the like. Further, mapping relation analysis is carried out on the expression data of the user through the target heart rate data. From the previously extracted target heart rate data and the corresponding facial expression data, the relationship and interaction between the two can be analyzed. For example, the target mapping relationship can be established by searching for facial expression distribution conditions under different concentricity levels through a statistical method or a machine learning algorithm. Based on the target mapping relationship, a mapping table is constructed to map the physiological state information and the facial information. And corresponding the characteristic values of the physiological state information and the facial information with the target mapping relation to generate a target mapping table. Thus, by inquiring the target mapping table, a corresponding mapping result can be obtained according to the physiological state information and the facial information of the target user and used for subsequent therapeutic instrument adjustment and personalized treatment. For example, assume that the server performs cluster analysis on physiological state information using a K-means clustering algorithm to obtain three clusters: A. b and C. And according to the physiological state information of the target user, the server extracts corresponding heart rate data from each cluster. For example, the physiological state information of the target user is categorized into cluster a, and the server extracts heart rate data matching the target user from the cluster. Meanwhile, the server extracts facial expression information of the target user, such as smile, anger, etc., using a facial expression recognition algorithm. Next, the server performs a mapping relationship analysis on the target heart rate data and the facial expression data. It is assumed that the server observes that the target user more easily reveals an angry facial expression in the high heart rate state and a smiling facial expression in the low heart rate state. Such observations may be made through statistical methods or machine learning algorithms. For example, the server analyzes the facial expression distribution at different heart rate levels in a large number of data sets and establishes a target mapping relationship, i.e. a relationship between heart rate level and facial expression. After the target mapping relation exists, the server builds a mapping table according to the current physiological state information and the facial information of the target user. By querying the target mapping table, the server maps the physiological state information and facial information of the target user to specific emotional states, such as happy, angry, etc. Thus, the server obtains the target mapping table, which contains the mapping relation between the physiological state information and the facial information and the corresponding emotion state.
In a specific embodiment, as shown in fig. 3, the process of executing step S104 may specifically include the following steps:
s301, performing noise processing on the face information to generate candidate face information;
s302, performing data scaling processing on the candidate face information to generate scaled face information;
s303, carrying out mirror symmetry processing on the scaled face information to generate extended face information.
The face information is subjected to noise processing. Noise processing is to reduce interference and unnecessary variance in the face information. Filters or other signal processing techniques may be used to smooth the facial information, removing high frequency noise or outliers. This allows clearer and more accurate face information to be obtained. Next, data scaling processing is performed on the face information after the noise processing. The scale of the face information can be adjusted by data scaling, so that the face information can adapt to different application scenes or analysis requirements. Common scaling methods include linear scaling, mean variance scaling, and the like. By the scaling processing, the numerical range of the face information can be mapped into a specified range for subsequent processing and analysis. Then, mirror symmetry processing is performed. The mirror symmetry process is to increase the diversity and variability of the face information. By left-right mirror-inverting the face information, a symmetrical face image can be generated. This can expand the data set of facial information, providing a wider variety of training samples or analysis samples. The mirror symmetry process may be achieved by horizontally flipping the facial image. For example, assume that the server has a facial image that includes a left side facial feature of a person. The server first performs noise processing on the image, smoothing out some fine texture or noise. Then, the server performs data scaling processing on the processed face image, scaling it to an appropriate size. The server horizontally inverts the original face image through mirror symmetry processing to generate a symmetrical right side face image. In this way, the server obtains an extended face information dataset comprising left and right side face features. Through the steps, the server successfully realizes noise processing, data scaling and mirror symmetry processing on the face information, and expanded face information is generated. Such processing may improve the quality and usability of the facial information, providing more accurate and comprehensive facial information data for subsequent analysis, recognition, or processing tasks.
In a specific embodiment, as shown in fig. 4, the process of executing step S106 may specifically include the following steps:
s401, carrying out data combination on the extended physiological state information and the working parameter information to generate a data set to be processed;
s402, performing decentralization processing on a data set to be processed to generate a zero-axis distribution data set;
s403, standard deviation calculation is carried out on the zero axis distribution data set, and corresponding standard deviation is generated;
s404, scaling the zero-axis distribution data set through standard deviation to generate a scaled data set;
s405, performing logarithmic transformation processing on the scaled data set to generate a corresponding training data set.
Specifically, the extended physiological state information and the working parameter information are subjected to data combination to generate a data set to be processed. This may be achieved by matching and merging the two data sets according to the respective identifiers or indexes. The combined data set contains the extended physiological state information and the working parameter information, and provides a unified data source for subsequent processing. Next, the data set to be processed is subjected to a decentralization process to generate a zero-axis distribution data set. The de-centering process may adjust the data distribution to be centered around zero by subtracting the mean of the data set to bring the mean of the data set close to zero. This eliminates the overall offset and trend of the data set, highlighting the relative changes in the data. And then, carrying out standard deviation calculation on the zero-axis distribution data set to generate a corresponding standard deviation. The standard deviation is an index for measuring the degree of data dispersion in the data set. By calculating the standard deviation of the zero-axis distribution data set, the degree of dispersion and the variation range of the data can be known. The zero-axis distribution dataset is then scaled by the standard deviation to generate a scaled dataset. Scaling may adjust the numerical range of the dataset to a particular scale or range. Common scaling methods include min-max scaling and normalized scaling. By scaling the data set, the magnitude of the data can be unified, eliminating the magnitude differences between different features. And carrying out logarithmic transformation processing on the scaled data set to generate a corresponding training data set. Logarithmic transformation can convert values in a dataset into their logarithmic values, common transformation methods include natural logarithms, logarithmic functions, and the like. The logarithmic transformation can reduce the skewness of the data and compress larger numerical value difference, so that the data more accords with the normal distribution assumption or meets the requirement of the linear relation. By way of example, assume that the server has an extended physiological state information dataset and an operating parameter information dataset. The server first merges the two data sets according to the corresponding identifiers to generate a data set to be processed. Then, the server performs a decentralization process on the data set to be processed, and the data distribution is made to be close to zero by subtracting the mean value of the data set. Next, the server calculates the standard deviation of the de-centralized dataset to obtain an index that measures the degree of data dispersion. And (3) performing scale scaling processing by the server through the standard deviation, and adjusting the data range to a specified scale. The server performs logarithmic transformation processing on the scaled data set to convert the data into logarithmic values. Therefore, the distribution form of the data can be further adjusted to be more in line with the normal distribution assumption or meet the requirement of the linear relation. The logarithmic transformation can reduce the skewness of the data, compress larger numerical differences, and make the data more comparable and interpretable. For example, assume that the server has an extended physiological state information data set containing blood pressure, blood glucose, heart rate, etc., and an operating parameter information data set containing parameters such as power, wavelength, and operating time. The server first combines the two data sets according to the corresponding identifiers to obtain a data set to be processed. Next, the server performs a de-centering process on the data set to be processed, and the data distribution is made to approach the zero axis by subtracting the mean value of the data set. The server then calculates the standard deviation of the de-centered dataset, resulting in a standard deviation for each feature. Taking a certain feature as an example, it is assumed that the standard deviation of the feature in the data set to be processed is 0.5. The server scales the data set according to this standard deviation, scaling the data to within 0.5 standard deviation centered around zero. The magnitude of the data can be unified, and magnitude differences among different features can be eliminated. The server performs a logarithmic transformation process on the scaled data set. Assuming that a feature has a value of 10 in the scaled dataset, the server converts it to a logarithmic value, e.g. taking the natural logarithm, the logarithmically transformed value is ln (10). Through the steps, the server performs data combination on the expanded physiological state information and the working parameter information, and performs decentralization, standard deviation calculation, scale scaling and logarithmic transformation processing on the data set to finally generate a training data set. Such training data sets may be used for training models, performing analysis and prediction tasks, and the like.
In a specific embodiment, the process of executing step S108 may specifically include the following steps:
(1) Collecting current facial information of a target user, extracting facial feature points of the current facial information through an intelligent regulation model of a target therapeutic instrument, and generating a plurality of facial feature points;
(2) Performing physiological state information mapping on the current facial information through a plurality of facial feature points to generate corresponding current physiological state information;
(3) And predicting the wavelength of the therapeutic instrument according to the current physiological state information to generate a target wavelength.
Specifically, the server collects current face information of the target user. This may be achieved by capturing and acquiring facial images using a suitable sensor or camera device. The sensor may record information such as the appearance characteristics, skin tone, and texture of the face. Next, facial feature point extraction is performed on the current facial information by the target therapeutic apparatus intelligent regulation model. This requires inputting a face image into a model using computer vision techniques, face recognition algorithms, and the like, to detect and extract the positions and shapes of the facial feature points. Facial feature points may include key points of eyes, nose, mouth, etc. The server maps physiological state information to the current facial information through a plurality of facial feature points. This process involves constructing a model of the association between facial feature points and physiological states. This may be trained by a machine learning method, and the model will learn how to map facial feature points onto corresponding physiological state information. For example, the server infers the heart rate, blood oxygen level, or emotional state of the user, etc., from the location and shape of the facial feature points. Once the server obtains current physiological state information, the server uses this information to make predictions of therapeutic device wavelength. This can be achieved by building a wavelength prediction model. The model may accept current physiological state information as input and output the appropriate therapeutic instrument wavelength as output. The predictive model may be based on a machine learning algorithm, such as a regression model, a decision tree, or a neural network, among others. For example, assume that a server is developing a facial diagnosis and treatment system that aims to predict the appropriate therapeutic instrument wavelength based on the facial information of the user. The server captures an image of the face of the target user using a high resolution camera. Through a pre-trained face detection model, the server extracts a plurality of facial feature points in the facial image, such as eyes, nose, mouth, and the like. The location and shape of these feature points will be used for subsequent physiological state information mapping. The server then trains using the existing data set to construct a model of the association between facial feature points and physiological state information. The model may predict a physiological state of the user, such as heart rate or emotional state, by the location and shape of facial feature points. The server inputs the predicted current physiological state information into a therapeutic instrument wavelength prediction model. The model may predict a target wavelength suitable for the user based on physiological state information of the user. The prediction model can learn the relation between the wavelength and the physiological state according to the existing training data, so that the most suitable therapeutic instrument wavelength can be accurately predicted according to the current physiological state.
In one embodiment, the process of performing therapeutic device wavelength prediction based on current physiological state information to generate the target wavelength may specifically include the following steps:
(1) Carrying out therapeutic instrument power data prediction on the LED photodynamic therapeutic instrument through the current physiological state information to generate predicted power data;
(2) Predicting the working time length of the LED photodynamic therapeutic instrument through the predicted power data to generate predicted working time length;
(3) And calculating the wavelength of the therapeutic instrument based on the predicted power data and the predicted working time length, and generating a target wavelength.
Specifically, current physiological state information of the target user, such as blood oxygen saturation, skin type, symptoms, etc., is collected. Such information may be obtained by means of sensors, medical devices or user filled questionnaires, etc. Next, a predictive model is built to predict power data of the therapeutic apparatus. The model may be based on a machine learning algorithm (e.g., regression model, neural network, etc.) or based on statistical analysis. The training data set of the model includes known physiological state information and corresponding therapeutic device power data. Through the training process of the model, the association relationship between the physiological state and the power can be established. And inputting the current physiological state information of the target user by using the trained model, and predicting the power data. The model will output corresponding predicted power data based on the input physiological state information. The predicted power data may be indicative of the power level that the therapeutic device should provide under a particular physiological condition. Next, the LED photodynamic therapy device is predicted for the operating time using the predicted power data. The prediction process considers the factors such as battery capacity and energy consumption of the therapeutic apparatus. By comparing the predicted power to the battery capacity of the therapeutic device, the operating time of the therapeutic device at a particular power can be estimated. For example, if the predicted power is 5 watts and the battery capacity of the therapeutic apparatus is 2000 milliamperes, then the predicted operating time period is 400 hours (assuming constant power). And calculating the wavelength of the therapeutic instrument based on the predicted power data and the predicted working time. Wavelength is an important parameter in phototherapy, and light of different wavelengths has different penetration depths and biological effects. According to the predicted power and working time, the design requirement of the therapeutic apparatus and the therapeutic effect target are combined, and the proper target wavelength can be determined. For example, if the predicted power is 5 watts, the predicted operating time period is 400 hours, and clinical studies have shown that light at a wavelength of 630 nanometers has the best therapeutic effect at such power and time period, then 630 nanometers may be selected as the target wavelength. In summary, the current physiological state information is used for carrying out power data prediction, working time prediction and target wavelength calculation on the LED photodynamic therapeutic apparatus, so that a personalized therapeutic scheme can be realized.
The method for intelligently adjusting the LED photodynamic therapy device according to the embodiment of the present invention is described above, and the intelligent adjusting system of the LED photodynamic therapy device according to the embodiment of the present invention is described below, referring to fig. 5, an embodiment of the intelligent adjusting system of the LED photodynamic therapy device according to the embodiment of the present invention includes:
the extraction module 501 is used for collecting working parameter information of the LED photodynamic therapeutic apparatus in the working process, extracting parameter dimensions of the working parameter information and generating a plurality of therapeutic apparatus parameter dimensions;
the acquisition module 502 is configured to acquire physiological status information of a target user, and simultaneously acquire facial information of the target user;
a construction module 503, configured to perform mapping table construction on the physiological status information and the facial information, and generate a target mapping table;
an expansion module 504, configured to perform data expansion on the face information, and generate expanded face information;
the mapping module 505 is configured to perform data mapping on the extended face information based on the target mapping table, and generate extended physiological status information;
a generating module 506, configured to perform standardization processing on the extended physiological status information and the working parameter information, and generate a training data set;
The training module 507 is configured to perform model training on a preset intelligent adjustment model of the initial therapeutic apparatus through the training data set, and perform parameter adjustment on the intelligent adjustment model of the initial therapeutic apparatus through a preset loss function, so as to obtain an intelligent adjustment model of the target therapeutic apparatus;
the input module 508 is configured to collect current face information of the target user, input the current face information to the target therapeutic apparatus intelligent adjustment model to perform therapeutic apparatus wavelength prediction, and generate a target wavelength;
the matching module 509 is configured to perform power matching on the LED photodynamic therapy device based on the target wavelength, generate a target power, and adjust the LED photodynamic therapy device based on the target wavelength and the target power.
Through the cooperation of the components, the working parameter information of the LED photodynamic therapeutic apparatus in the working process is collected, parameter dimension extraction is carried out on the working parameter information, a plurality of therapeutic apparatus parameter dimensions are generated, physiological state information of a target user is collected, and meanwhile face information of the target user is collected; building a mapping table of the physiological state information and the facial information to generate a target mapping table; performing data expansion on the face information to generate expanded face information; performing data mapping on the extended face information based on the target mapping table to generate extended physiological state information; carrying out standardized processing on the extended physiological state information and the working parameter information to generate a training data set; model training is carried out on a preset intelligent regulation model of the initial therapeutic apparatus through a training data set, and parameter regulation is carried out on the intelligent regulation model of the initial therapeutic apparatus through a preset loss function, so that an intelligent regulation model of the target therapeutic apparatus is obtained; collecting current face information of a target user, inputting the current face information into an intelligent regulation model of a target therapeutic instrument to predict the wavelength of the therapeutic instrument, and generating a target wavelength; and performing power matching on the LED photodynamic therapeutic instrument based on the target wavelength to generate target power, and adjusting the LED photodynamic therapeutic instrument based on the target wavelength and the target power. According to the application, the optimal treatment wavelength and power parameters are obtained by collecting physiological state information, facial information and working parameter information of the LED photodynamic therapeutic apparatus of a target user and predicting the LED wavelength based on the prediction model, and a series of parameters such as facial information, physiological state and the like are fused together by establishing the intelligent regulation model of the target therapeutic apparatus to form an intelligent therapeutic apparatus regulating system, so that the intelligent degree of the therapeutic apparatus is improved. By carrying out data expansion and mapping table construction of physiological state information on the facial information, a more complete and comprehensive training data set can be obtained, and the prediction accuracy and robustness of the model are improved. The physiological state information and the working parameter information are subjected to standardized processing, so that the data among different characteristics can be coordinated and consistent, and the processibility of the data and the stability of the model are improved. The LED photodynamic therapeutic instrument is adjusted based on the target wavelength and the power, the self-adaptive adjustment can be performed according to facial features, physiological states and photodynamic therapeutic requirements of different patients, and the accuracy and the efficiency of intelligent adjustment of the LED photodynamic therapeutic instrument are improved.
Fig. 5 above describes the intelligent adjusting system of the LED photodynamic therapy apparatus in the embodiment of the present invention in detail from the point of view of modularized functional entities, and the intelligent adjusting device of the LED photodynamic therapy apparatus in the embodiment of the present invention is described in detail from the point of view of hardware processing.
Fig. 6 is a schematic structural diagram of an intelligent adjusting apparatus of an LED photodynamic therapy apparatus according to an embodiment of the present invention, where the intelligent adjusting apparatus 600 of the LED photodynamic therapy apparatus may have relatively large differences according to different configurations or performances, and may include one or more processors (CPU) 610 (e.g., one or more processors) and a memory 620, and one or more storage mediums 630 (e.g., one or more mass storage devices) storing application programs 633 or data 632. Wherein the memory 620 and the storage medium 630 may be transitory or persistent storage. The program stored in the storage medium 630 may include one or more modules (not shown), each of which may include a series of instruction operations in the intelligent regulation device 600 of the LED photodynamic therapy apparatus. Still further, the processor 610 may be configured to communicate with the storage medium 630 to execute a series of instruction operations in the storage medium 630 on the intelligent regulation device 600 of the LED photodynamic therapy apparatus.
The intelligent regulation device 600 of the LED photodynamic therapy apparatus may also include one or more power supplies 640, one or more wired or wireless network interfaces 650, one or more input-output interfaces 660, and/or one or more operating systems 631, such as WindowsServe, macOSX, unix, linux, freeBSD, and the like. It will be appreciated by those skilled in the art that the smart tuning device structure of the LED photodynamic therapy device shown in fig. 6 does not constitute a limitation of the smart tuning device of the LED photodynamic therapy device, and may include more or fewer components than shown, or may combine certain components, or may be arranged in different components.
The invention also provides intelligent adjusting equipment of the LED photodynamic therapy apparatus, which comprises a memory and a processor, wherein the memory stores computer readable instructions, and when the computer readable instructions are executed by the processor, the processor executes the steps of the intelligent adjusting method of the LED photodynamic therapy apparatus in the above embodiments.
The present invention also provides a computer readable storage medium, which may be a non-volatile computer readable storage medium, and may also be a volatile computer readable storage medium, where instructions are stored in the computer readable storage medium, where the instructions when executed on a computer cause the computer to perform the steps of the intelligent adjustment method of the LED photodynamic therapy apparatus.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, which are not repeated herein.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a storage medium, including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a read-only memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, or other various media capable of storing program codes.
The above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. The intelligent adjusting method of the LED photodynamic therapeutic apparatus is characterized by comprising the following steps of:
collecting working parameter information of the LED photodynamic therapeutic instrument in the working process, extracting parameter dimensions of the working parameter information, and generating a plurality of therapeutic instrument parameter dimensions;
collecting physiological state information of a target user, and collecting facial information of the target user at the same time;
building a mapping table of the physiological state information and the facial information to generate a target mapping table;
performing data expansion on the face information to generate expanded face information;
performing data mapping on the extended face information based on the target mapping table to generate extended physiological state information;
Carrying out standardization processing on the extended physiological state information and the working parameter information to generate a training data set;
model training is carried out on a preset intelligent regulation model of the initial therapeutic apparatus through the training data set, and parameter regulation is carried out on the intelligent regulation model of the initial therapeutic apparatus through a preset loss function, so that an intelligent regulation model of the target therapeutic apparatus is obtained;
collecting current facial information of the target user, inputting the current facial information into the intelligent regulation model of the target therapeutic instrument to predict the wavelength of the therapeutic instrument, and generating a target wavelength;
and performing power matching on the LED photodynamic therapeutic instrument based on the target wavelength to generate target power, and adjusting the LED photodynamic therapeutic instrument based on the target wavelength and the target power.
2. The intelligent adjustment method of the LED photodynamic therapy apparatus according to claim 1, wherein the collecting the working parameter information of the LED photodynamic therapy apparatus during the working process, extracting the parameter dimension of the working parameter information, generating a plurality of therapeutic apparatus parameter dimensions, includes:
collecting power data of the LED photodynamic therapeutic instrument to generate power data;
Collecting wavelength data of the LED photodynamic therapeutic instrument to generate wavelength data;
collecting working time of the LED photodynamic therapeutic instrument to generate time data;
performing parameter integration on the power data, the wavelength data and the time data to generate working parameter data;
traversing the working parameter data parameter dimension to generate a plurality of therapeutic instrument parameter dimensions.
3. The method for intelligently adjusting an LED photodynamic therapy device according to claim 1, wherein the mapping table construction is performed on the physiological status information and the face information to generate a target mapping table, comprising:
performing cluster analysis on the physiological state information to generate a corresponding cluster feature set;
extracting heart rate data of the target user through the cluster feature set to generate target heart rate data;
extracting the facial information by using the user expression to generate corresponding user expression information;
carrying out mapping relation analysis on the user expression information through the target heart rate data to generate a corresponding target mapping relation;
and constructing a mapping table of the physiological state information and the facial information through the target mapping relation to generate a target mapping table.
4. The intelligent adjustment method of the LED photodynamic therapy device according to claim 1, wherein the data expansion of the face information to generate expanded face information includes:
performing noise processing on the face information to generate candidate face information;
performing data scaling processing on the candidate face information to generate scaled face information;
and carrying out mirror symmetry processing on the scaled face information to generate extended face information.
5. The method for intelligently adjusting the LED photodynamic therapy device according to claim 1, wherein the step of normalizing the extended physiological state information and the operating parameter information to generate a training data set includes:
carrying out data combination on the extended physiological state information and the working parameter information to generate a data set to be processed;
performing decentration on the data set to be processed to generate a zero-axis distribution data set;
performing standard deviation calculation on the zero-axis distribution data set to generate a corresponding standard deviation;
scaling the zero-axis distribution data set by the standard deviation to generate a scaled data set;
and carrying out logarithmic transformation processing on the scaled data set to generate a corresponding training data set.
6. The method for intelligently adjusting an LED photodynamic therapy device according to claim 1, wherein the steps of collecting current face information of the target user, inputting the current face information to the intelligent adjustment model of the target photodynamic therapy device to predict a wavelength of the therapy device, and generating a target wavelength include:
collecting current facial information of the target user, extracting facial feature points of the current facial information through the intelligent regulation model of the target therapeutic instrument, and generating a plurality of facial feature points;
performing physiological state information mapping on the current facial information through the facial feature points to generate corresponding current physiological state information;
and predicting the wavelength of the therapeutic instrument according to the current physiological state information to generate a target wavelength.
7. The method of claim 6, wherein the predicting the therapeutic wavelength based on the current physiological state information to generate the target wavelength comprises:
carrying out therapeutic instrument power data prediction on the LED photodynamic therapeutic instrument through the current physiological state information to generate predicted power data;
predicting the working time length of the LED photodynamic therapeutic instrument according to the predicted power data to generate predicted working time length;
And calculating the wavelength of the therapeutic instrument based on the predicted power data and the predicted working time length, and generating a target wavelength.
8. An intelligent regulation system of an LED photodynamic therapy apparatus, characterized in that the intelligent regulation system of the LED photodynamic therapy apparatus comprises:
the extraction module is used for collecting working parameter information of the LED photodynamic therapeutic instrument in the working process, extracting parameter dimensions of the working parameter information and generating a plurality of therapeutic instrument parameter dimensions;
the acquisition module is used for acquiring physiological state information of a target user and acquiring facial information of the target user at the same time;
the construction module is used for constructing a mapping table of the physiological state information and the facial information and generating a target mapping table;
the expansion module is used for carrying out data expansion on the face information to generate expanded face information;
the mapping module is used for carrying out data mapping on the extended face information based on the target mapping table to generate extended physiological state information;
the generating module is used for carrying out standardized processing on the extended physiological state information and the working parameter information to generate a training data set;
the training module is used for carrying out model training on a preset intelligent regulation model of the initial therapeutic apparatus through the training data set, and carrying out parameter regulation on the intelligent regulation model of the initial therapeutic apparatus through a preset loss function to obtain an intelligent regulation model of the target therapeutic apparatus;
The input module is used for acquiring the current facial information of the target user, inputting the current facial information into the intelligent regulation model of the target therapeutic instrument to predict the wavelength of the therapeutic instrument and generating a target wavelength;
and the matching module is used for carrying out power matching on the LED photodynamic therapeutic instrument based on the target wavelength, generating target power and adjusting the LED photodynamic therapeutic instrument based on the target wavelength and the target power.
9. An intelligent regulation equipment of LED photodynamic therapy appearance, its characterized in that, the intelligent regulation equipment of LED photodynamic therapy appearance includes: a memory and at least one processor, the memory having instructions stored therein;
the at least one processor invokes the instructions in the memory to cause the intelligent regulation device of the LED photodynamic therapy instrument to perform the intelligent regulation method of the LED photodynamic therapy instrument as claimed in any one of claims 1 to 7.
10. A computer readable storage medium having instructions stored thereon, wherein the instructions when executed by a processor implement the intelligent regulation method of the LED photodynamic therapy device according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310725849.4A CN116440425B (en) | 2023-06-19 | 2023-06-19 | Intelligent adjusting method and system of LED photodynamic therapeutic instrument |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310725849.4A CN116440425B (en) | 2023-06-19 | 2023-06-19 | Intelligent adjusting method and system of LED photodynamic therapeutic instrument |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116440425A CN116440425A (en) | 2023-07-18 |
| CN116440425B true CN116440425B (en) | 2023-08-18 |
Family
ID=87122372
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310725849.4A Active CN116440425B (en) | 2023-06-19 | 2023-06-19 | Intelligent adjusting method and system of LED photodynamic therapeutic instrument |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116440425B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117101017B (en) * | 2023-10-08 | 2024-09-03 | 广州中科智慧健康产业科技有限公司 | Personalized phototherapy room adjusting method and system based on biofeedback |
| CN117323578B (en) * | 2023-12-01 | 2024-01-30 | 深圳恩多克医疗有限公司 | Intelligent control method and system for infrared therapeutic apparatus |
| CN117339118B (en) * | 2023-12-05 | 2024-03-12 | 深圳恩多克医疗有限公司 | Intelligent monitoring and controlling method and system for illumination intensity of infrared therapeutic apparatus |
| CN118001597B (en) * | 2024-03-12 | 2024-08-13 | 丰城市人民医院 | Control error regulation and control system and method for tumor therapeutic apparatus |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008062000A1 (en) * | 2006-11-20 | 2008-05-29 | Spectracure Ab | System and method for predicting and/or adjusting control parameters of interstitial photodynamic light therapy |
| CN112101203A (en) * | 2020-09-15 | 2020-12-18 | 黑龙江登源科技有限公司 | Facial beauty medical design system and method based on data analysis |
| CN112364817A (en) * | 2020-11-26 | 2021-02-12 | 深圳市旗华科技有限公司 | AI control system based on facial expression psychophysiological parameters |
| CN112741593A (en) * | 2019-10-29 | 2021-05-04 | 重庆医科大学附属儿童医院 | Pain assessment method, system, device and storage medium |
| CN113140312A (en) * | 2020-01-19 | 2021-07-20 | Oppo广东移动通信有限公司 | User data processing method and device, session data processing method and device, and electronic equipment |
| CN113762459A (en) * | 2021-01-26 | 2021-12-07 | 北京沃东天骏信息技术有限公司 | Model training method, text generation method, device, medium and equipment |
| JP2022029219A (en) * | 2020-08-04 | 2022-02-17 | レナード株式会社 | Beauty support system, and beauty support method |
| CN114764784A (en) * | 2021-01-04 | 2022-07-19 | 深圳科亚医疗科技有限公司 | Training method and system of machine learning model for physiological relevant parameter prediction |
| CN115531720A (en) * | 2022-10-20 | 2022-12-30 | 山东丝琳医药科技有限公司 | Skin physical therapy device, equipment and parameter adjusting method thereof |
| CN116035553A (en) * | 2023-01-31 | 2023-05-02 | 西安电子科技大学 | Human body composite parameter measurement and photoelectric treatment integrated diagnosis and treatment equipment |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11526808B2 (en) * | 2019-05-29 | 2022-12-13 | The Board Of Trustees Of The Leland Stanford Junior University | Machine learning based generation of ontology for structural and functional mapping |
-
2023
- 2023-06-19 CN CN202310725849.4A patent/CN116440425B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008062000A1 (en) * | 2006-11-20 | 2008-05-29 | Spectracure Ab | System and method for predicting and/or adjusting control parameters of interstitial photodynamic light therapy |
| CN112741593A (en) * | 2019-10-29 | 2021-05-04 | 重庆医科大学附属儿童医院 | Pain assessment method, system, device and storage medium |
| CN113140312A (en) * | 2020-01-19 | 2021-07-20 | Oppo广东移动通信有限公司 | User data processing method and device, session data processing method and device, and electronic equipment |
| JP2022029219A (en) * | 2020-08-04 | 2022-02-17 | レナード株式会社 | Beauty support system, and beauty support method |
| CN112101203A (en) * | 2020-09-15 | 2020-12-18 | 黑龙江登源科技有限公司 | Facial beauty medical design system and method based on data analysis |
| CN112364817A (en) * | 2020-11-26 | 2021-02-12 | 深圳市旗华科技有限公司 | AI control system based on facial expression psychophysiological parameters |
| CN114764784A (en) * | 2021-01-04 | 2022-07-19 | 深圳科亚医疗科技有限公司 | Training method and system of machine learning model for physiological relevant parameter prediction |
| CN113762459A (en) * | 2021-01-26 | 2021-12-07 | 北京沃东天骏信息技术有限公司 | Model training method, text generation method, device, medium and equipment |
| CN115531720A (en) * | 2022-10-20 | 2022-12-30 | 山东丝琳医药科技有限公司 | Skin physical therapy device, equipment and parameter adjusting method thereof |
| CN116035553A (en) * | 2023-01-31 | 2023-05-02 | 西安电子科技大学 | Human body composite parameter measurement and photoelectric treatment integrated diagnosis and treatment equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116440425A (en) | 2023-07-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116440425B (en) | Intelligent adjusting method and system of LED photodynamic therapeutic instrument | |
| Khan et al. | Biometric systems utilising health data from wearable devices: applications and future challenges in computer security | |
| WO2017193497A1 (en) | Fusion model-based intellectualized health management server and system, and control method therefor | |
| US10825569B2 (en) | Universal non-invasive blood glucose estimation method based on time series analysis | |
| US9053308B2 (en) | Multi electro-biometric user recognition | |
| US20070179367A1 (en) | Method and Apparatus for Noninvasively Estimating a Property of an Animal Body Analyte from Spectral Data | |
| CN109568760A (en) | Sleep environment adjusting method and system | |
| CN117598700B (en) | Intelligent blood oxygen saturation detection system and method | |
| Chen et al. | Machine learning method for continuous noninvasive blood pressure detection based on random forest | |
| CN107803843A (en) | A kind of measuring of human health robot based on Raspberry Pi | |
| CN117617921B (en) | Intelligent blood pressure monitoring system and method based on Internet of things | |
| CN109998497B (en) | Sleep-in detection and judgment system in luminous environment | |
| CN115952460A (en) | Emotion recognition method and device, computer equipment and storage medium | |
| CN118248332A (en) | Intelligent multi-mode physiological data monitoring system | |
| CN116392085A (en) | Sleep stability quantification and adjustment method, system and device based on trend analysis | |
| KR102108961B1 (en) | An image-based artificial intelligence deep-learning based blood pressure estimation apparatus | |
| CN115089179A (en) | Psychological emotion insights analysis method and system | |
| CN111281403B (en) | Fine-grained human body fatigue detection method and device based on embedded equipment | |
| CN117860207B (en) | Video non-contact measurement pain identification method and system based on data analysis | |
| CN110135357B (en) | Happiness real-time detection method based on remote sensing | |
| CN116530981A (en) | Facial recognition-based qi and blood state analysis system and method | |
| CN115317304A (en) | Intelligent massage method and system based on physiological characteristic detection | |
| CN116434979A (en) | Physiological state cloud monitoring method, monitoring system and storage medium | |
| KR20200057666A (en) | Method for analyze user's skin and construct bigdata using skin measurement and care process | |
| CN110693508A (en) | Multi-channel cooperative psychophysiological active sensing method and service robot |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |