WO2020010668A1 - 基于睡眠大数据的人体健康评估方法及评估*** - Google Patents

基于睡眠大数据的人体健康评估方法及评估*** Download PDF

Info

Publication number
WO2020010668A1
WO2020010668A1 PCT/CN2018/100955 CN2018100955W WO2020010668A1 WO 2020010668 A1 WO2020010668 A1 WO 2020010668A1 CN 2018100955 W CN2018100955 W CN 2018100955W WO 2020010668 A1 WO2020010668 A1 WO 2020010668A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
sleep
training
human
human body
Prior art date
Application number
PCT/CN2018/100955
Other languages
English (en)
French (fr)
Inventor
杨向东
单华锋
曹辉
宋琪隆
李红文
朱震宇
韩秀萍
Original Assignee
浙江清华长三角研究院
麒盛科技股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 浙江清华长三角研究院, 麒盛科技股份有限公司 filed Critical 浙江清华长三角研究院
Priority to RU2020129796A priority Critical patent/RU2757048C1/ru
Priority to AU2018432124A priority patent/AU2018432124A1/en
Priority to EP18926215.7A priority patent/EP3783619A4/en
Priority to US17/040,975 priority patent/US20210225510A1/en
Publication of WO2020010668A1 publication Critical patent/WO2020010668A1/zh

Links

Images

Classifications

    • 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
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
    • 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
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/0002—Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network
    • A61B5/0015—Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network characterised by features of the telemetry system
    • A61B5/0022—Monitoring a patient using a global network, e.g. telephone networks, internet
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
    • A61B5/0205—Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
    • A61B5/024—Detecting, measuring or recording pulse rate or heart rate
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/08—Detecting, measuring or recording devices for evaluating the respiratory organs
    • A61B5/0816—Measuring devices for examining respiratory frequency
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
    • A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
    • A61B5/1116—Determining posture transitions
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
    • A61B5/6887—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient mounted on external non-worn devices, e.g. non-medical devices
    • A61B5/6892—Mats
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235—Details of waveform analysis
    • A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
    • A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
    • G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H15/00—ICT specially adapted for medical reports, e.g. generation or transmission thereof
    • 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
    • G16H40/67—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 for remote operation
    • 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
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients

Definitions

  • the invention relates to the technical field of sleep health data analysis and management, in particular to a human health evaluation method and evaluation system based on sleep big data.
  • the commonly used methods of physical health assessment include questionnaire surveys, testing of various physical indicators with medical equipment, daily wearing of wearable devices with physiological data for detection, and the like.
  • the questionnaire survey is a health assessment questionnaire provided by users participating in the survey for medical staff, and combined with the physical state of daily life to answer the questionnaire, the medical staff collects the questionnaire filled out by the user to evaluate the user's health status.
  • To detect physiological data through medical equipment or wearable equipment it is necessary to regularly record the test results, and timely discover health problems based on the recorded test data.
  • the data sources used in these assessment methods are not very authentic, and are easily affected by user consciousness during the data acquisition process. It is difficult to completely record the physiological data in normal living habits, and the standard and accuracy of data collection cannot be guaranteed.
  • Sleep accounts for 1/3 of people's daily life.
  • the physiological data of the human body in the sleep state is relatively stable. Collecting the physiological data of the human body in the sleep state for health assessment can ensure the accuracy, stability and durability of the data. Assess the accuracy of the results.
  • the medically accepted method for judging the sleep state in clinical practice is the sleep polygraph method.
  • brain waves, eye movements, and Ottoelectricity can be measured to determine sleep status from these waveforms.
  • methods such as electroencephalogram and tendon measurement require electrodes to be placed on the subject's body, and the burden on the subject is very large. Therefore, it is impossible to measure brain waves, tendon electricity, and the like in ordinary households.
  • Physical assessment is an objective and systematic data collection process. Based on health history and head-to-toe or general system checks, good physical assessment results require an organized, systematic approach that directly responds to users. The exploration and discovery of actual or potential problems and research. Therefore, how to ensure that the user's physiological data collection in the early period is sufficient and comprehensive enough, how to ensure the rationality of data processing and reduce errors, is a great concern for human health assessment. problem.
  • An object of the present invention is to provide a human health evaluation method and evaluation system based on sleep big data in view of the shortcomings in the prior art.
  • the health big data research in this system refers to the long-term, stable, and continuous growth of human life characteristics data, and the use of advanced data mining technology to clearly and clearly describe the physical status of human individuals and predict their healthy future.
  • the model automatically learns features related to human health, thereby saving the time cost of manual selection and construction of features, achieving the goal of accurately predicting disease, and further training
  • the classifier model can directly know the type of disease in the health assessment report.
  • the technical solution adopted by the present invention is: a human health assessment method based on sleep big data, including the following steps:
  • the above-mentioned sleep data stored in the cloud server is subjected to data pre-processing to filter out missing and erroneous data, and the correct data is put into the artificial intelligence learning model for training, so that the artificial intelligence learning model learns the disease characteristics and calculates the physiology Evaluation index
  • a human health status analysis report is generated based on the human physiological evaluation indexes and the classifier model obtained through the training of the above data.
  • the correct data after the preprocessing in the above method is divided into sleep data with diseased tags and sleep data without diseased tags, and the sleep data with diseased tags comes from known Diseased type of human body, and the disease-free label sleep data comes from a human body of unknown diseased state.
  • the process of putting the above correct data into an artificial intelligence learning model for training includes:
  • the unsupervised network loss is calculated by sending the pre-processed sleep data without the diseased label into the self-encoding network;
  • the above-mentioned dimensionality-reduced output data of the sleep data with the diseased label in the self-encoding network is input to a classifier to train a classifier model.
  • the physiological evaluation index is used to evaluate whether a human body is in a diseased state
  • the classifier model is used to analyze a human body diseased type.
  • a human health assessment system based on sleep big data the system includes:
  • a sleep data acquisition unit configured to acquire various physiological data of the human body during sleep, the data coming from sensors installed on the bed;
  • a sleep data storage unit including a cloud server, for storing all sleep data collected by the sleep data acquisition unit;
  • Data training unit which is used to pre-process sleep data and train the data through artificial intelligence learning models to obtain physiological evaluation indicators
  • Classifier model training unit for classifier model training on sleep data after training of an artificial intelligence learning model
  • the report generating unit is configured to generate a human health status analysis report according to the human physiological evaluation index and the classifier model obtained through data training.
  • the data training unit includes:
  • Data pre-processing used to filter out missing and incorrect sleep data stored in cloud servers
  • a self-encoder is used to create a self-encoding network and iteratively train the data by calculating the network loss.
  • the technical solution adopted by the present invention is further: a terminal device, the device including the above-mentioned human health evaluation system based on sleep big data.
  • the device further includes an operation mechanism for controlling the report generation unit to generate a report.
  • the system of the present invention realizes data transmission with sensors installed on the bed.
  • the bed has a long service life and the user's various physiological data in the sleep state are relatively stable, thereby ensuring the durability and stability of data acquisition.
  • the massive sleep data obtained by the installed sensors provides a high-quality data basis for health assessment.
  • the cloud server of the system of the present invention is used to store the daily sleep data of the user, which ensures the stability and security of the sleep data, and also provides a large amount of data foundation for the health evaluation, and improves the accuracy of the evaluation results.
  • the method of the present invention acquires a large amount of physiological data of the human body during sleep, and performs artificial intelligence learning model training, and the person's active consciousness is very weak in the sleep state.
  • the physiological data obtained at this time can truly reflect the health status of the human body;
  • the applicable artificial intelligence learning model allows the model to automatically learn the features related to human health, thereby saving the time cost of manually selecting and constructing features, and achieving the purpose of accurately predicting diseases.
  • FIG. 1 is a schematic flowchart of a human health assessment method based on sleep big data according to an embodiment of the present invention
  • FIG. 2 is a schematic flowchart of data training of an artificial intelligence learning model according to an embodiment of the present invention
  • FIG. 3 is a schematic flowchart of classifier model training according to an embodiment of the present invention.
  • FIG. 4 is a framework diagram of a human health evaluation system based on sleep big data according to an embodiment of the present invention.
  • FIG. 1 shows a flowchart of a human health assessment method based on sleep big data in an embodiment of the present invention.
  • a human health assessment method based on sleep big data may include:
  • Step 101 Through the sensors installed on the bed, various physiological data of the human body during sleep are acquired, and these sleep data are permanently stored in the cloud server.
  • the human health assessment method based on sleep big data in the embodiment of the present invention is a method of training data by using an artificial intelligence learning model to process, analyze, and analyze a large amount of sleep data stored in a cloud server. training.
  • the method realizes data transmission through a cloud server and sensors installed on the bed.
  • the cloud server can also communicate with the database of the control system to obtain real-time and stable user sleep data.
  • the various types of physiological data stored in the cloud server during human sleep are associated with the types of sensors installed on the bed. In actual applications, specific sensors can be selected to obtain target data based on the focus of the health assessment, such as the physiological data obtained in this embodiment. Data include: heart rate, respiration rate, turning over, fretting, snoring, and out-of-bed information.
  • Step 102 Perform the data preprocessing on the sleep data stored in the cloud server to filter out missing and erroneous data, put the correct data into the artificial intelligence learning model for training, and make the artificial intelligence learning model learn the disease characteristics. Calculate the physiological evaluation index.
  • the original human sleep data stored in the cloud server is obtained in real time without any processing, and it will inevitably contain a large amount of missing and erroneous data. Therefore, before data analysis, data must be pre-processed and the correct sleep data can be filtered to be stored Training in artificial intelligence learning models, otherwise the whole model will be disordered, and correct prediction results cannot be obtained.
  • the data preprocessing divides the sleep data into the sleep data with the diseased label and the sleep data without the diseased label according to the data source while filtering out the noise data.
  • the sleep data with the diseased label refers to the data collected from the known disease. Types of human physiological data; sleep data without disease tags are collected from human physiological data of unknown disease state.
  • the pre-processed sleep data will be sent to the artificial intelligence learning model for training and can be used to evaluate the health status of the human body.
  • the sleep data when collecting sleep data with disease labels for artificial intelligence learning model training and classifier model training, an individual who has accurately learned the disease state and knows the disease type will be selected.
  • the sleep data is collected and input into the artificial intelligence learning model, which is used to make the model recognize whether the human body is normal or diseased and the type of disease.
  • sleep-free label sleep data for artificial intelligence learning model training an individual who is not sure whether or not he or she is sick will be selected, and his sleep data will be collected and entered into the artificial intelligence learning model summary for the model to learn about human sleep What the data has in common.
  • Step 103 Use the diseased features of the data learned by the artificial intelligence learning model to train a classifier model, so that the classifier model can identify the diseased types corresponding to different data.
  • the data trained by the artificial intelligence learning model will be used as a physiological evaluation index, which can be used to identify whether an individual's physical state is a diseased state or a normal state.
  • the sleep data with the diseased label is passed through the dimensionality reduction output of the self-encoding network as the input of the SVM classifier to train the classifier model so that the classifier model can identify the diseases corresponding to different data. Disease type.
  • the artificial intelligence learning model learns some characteristics related to the disease from the original data, and then further training through step 103 enables the classifier model to identify the specific type of disease suffered by the individual who produced the sleep data.
  • Step 104 Generate a human health status analysis report according to the human physiological evaluation index and the classifier model obtained through the training of the data.
  • the sleep data stored in the cloud server is pre-processed to remove missing and erroneous data.
  • the remaining correct sleep data is divided into sleep data with diseased tags and sleep data without diseased tags according to the data source. , Put these data into the artificial intelligence learning model for training.
  • FIG. 2 shows a schematic flowchart of data training of an artificial intelligence learning model according to an embodiment of the present invention. As shown in FIG. 2, the correct data after preprocessing is put into the artificial intelligence learning model for training, including:
  • the pre-processed sleep data x without the diseased label is sent to a deep autoencoder of the self-encoding network, and the unsupervised network loss loss1 k is calculated.
  • the optimal calculation method is selected as follows: the sleep data x is sent to a depth autoencoder to obtain the mean square error value of the input and output, and the value is given to the unsupervised network loss loss1 k .
  • the pre-processed sleep data y with the diseased label is sent to the self-encoding network with the same parameters as above, to obtain the dimensionality reduction output data of the middle layer and calculate the loss 2 k of the supervised network loss.
  • the present application is calculated for the optimum network supervised way for loss loss2 k: the intermediate layer into output data dimension reduction obtained diseased tag different classes, each class to find the center point, and the same data to calculate from the same center point, and given the data obtained from supervised network losses loss2 k.
  • the neural network loss and self-encoded network loss obtained from the training will be used as physiological evaluation indicators to evaluate whether the individual corresponding to the sleep data is in a sick state or a normal healthy state.
  • the middle-layer dimensionality-reduced output data obtained when the above-mentioned sickly labeled sleep data y is sent to the self-encoding network will be further used as the input of the SVM classifier to train the classifier model so that the classifier model can identify The type of illness corresponding to different data.
  • the trained classifier model can identify the specific type of disease suffered by the individual who produced the sleep data.
  • the physiological evaluation indexes and classifier models obtained through the above training can be used to generate an individual health status analysis report corresponding to sleep data.
  • the sleep data of individuals who need to be evaluated for health are selected from the cloud server, and these data are put into the artificial intelligence learning model that has been trained, and the disease state of the individual is determined by the model. If the result is The disease state can further obtain the type of disease of the individual.
  • a human health evaluation report containing the above evaluation results can be output or printed by the terminal device, so as to achieve the purpose of using sleep data to predict diseases.
  • FIG. 4 is a framework diagram of a human health assessment system based on sleep big data according to an embodiment of the present invention. As shown in FIG. 4, the human health assessment system based on sleep big data includes:
  • a sleep data acquisition unit configured to acquire various physiological data of the human body during sleep, the data coming from sensors installed on the bed;
  • a sleep data storage unit including a cloud server, for storing all sleep data collected by the sleep data acquisition unit;
  • the data training unit is used for preprocessing the sleep data, and training the data through an artificial intelligence learning model to obtain physiological evaluation indexes.
  • the data training unit includes: data preprocessing for filtering out missing and erroneous sleep data stored in the cloud server; self-encoders for creating self-encoding networks, and iteratively training the data by calculating network losses.
  • Classifier model training unit for classifier model training on sleep data after training of an artificial intelligence learning model
  • the report generating unit is configured to generate a human health status analysis report according to the human physiological evaluation index and the classifier model obtained through data training.
  • the system of the present invention can be set in a control system of an electric bed or in a terminal device for human-computer interactive operation.
  • the device further includes an operating mechanism for controlling a report generating unit to generate a report, so that the user can follow Need to customize the evaluation system.
  • the terminal devices described in the embodiments of the present specification may include Internet devices such as user equipment, smart phones, computers, mobile Internet devices, or wearable smart devices, which are not limited in the embodiments of the present invention.
  • the embodiments of the present invention may be provided as a method, a system, or a computer program product. Therefore, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects.

Landscapes

  • Health & Medical Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Medical Informatics (AREA)
  • Public Health (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • General Health & Medical Sciences (AREA)
  • Pathology (AREA)
  • Epidemiology (AREA)
  • Primary Health Care (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • Databases & Information Systems (AREA)
  • Molecular Biology (AREA)
  • Veterinary Medicine (AREA)
  • Animal Behavior & Ethology (AREA)
  • Surgery (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Biophysics (AREA)
  • Physiology (AREA)
  • Cardiology (AREA)
  • Artificial Intelligence (AREA)
  • Pulmonology (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Evolutionary Computation (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Theoretical Computer Science (AREA)
  • Dentistry (AREA)
  • General Business, Economics & Management (AREA)
  • Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Signal Processing (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Psychiatry (AREA)
  • Fuzzy Systems (AREA)
  • Medical Treatment And Welfare Office Work (AREA)

Abstract

一种基于睡眠大数据的人体健康评估方法及评估***，该方法和***长期地、稳定地、持续增长地获得人体生命特征数据，采用先进的数据挖掘技术清晰明确地描述人类个体的身体状态并预测其健康的未来。通过安装在床上的传感器，获取人体睡眠时的各项生理数据存储于云端服务器（101）；运用人工智能学习模型得到人体生理评价指标并生成人体健康状态分析报告。通过获取安装于床上的传感器数据，进行人工智能学习模型数据训练，让模型自动学习出与人体健康相关的特征，从而节省人工挑选和构造特征的时间成本，达到精准预测疾病的目的，进一步通过训练分类器模型可以在健康评估报告中直接获知患病类型。

Description

基于睡眠大数据的人体健康评估方法及评估*** 技术领域
本发明涉及睡眠健康数据分析与管理技术领域，具体地说，是一种基于睡眠大数据的人体健康评估方法及评估***。
背景技术
目前人们常用的身体健康评估方法主要包括问卷调查、借助医疗器械测试身体各项指标、日常佩戴具有检测生理数据的穿戴设备等等。其中，问卷调查是参与调查的用户针对医护人员提供的健康评估问卷，结合日常生活中身体状态回答问卷，医护人员收集用户填写的调查问卷后对用户健康状态进行评估。通过医疗器械或穿戴设备检测生理数据，需要定期记录检测结果，根据记录的检测数据及时发现健康问题。这些评估方法所使用的数据来源真实性不高，数据获取过程中很容易受到用户意识的影响，很难完整记录正常生活习惯中的生理数据，无法保证数据采集的标准性和精确度。
睡眠占人们日常生活1/3的时间，睡眠状态下人体各项生理数据相对稳定，采集睡眠状态下的人体生理数据用来进行健康评估，可以保证数据的准确性、稳定性和持久性，提供评估结果的准确度。现在医学上被认可的判断临床应用的睡眠状态的方法是睡眠聚图法。在睡眠聚图法中，测量脑波、眼球运动和奥特筋电，可以从这些波形中判断睡眠状态。然而，测量脑波、筋电等的方法，需要在被试验者的身体上安装电极，对实验者的负担非常大。因此，不可能在一般家庭中测量脑波和筋电等。
身体评估是一个有组织的***性的收集过程所产生的客观数据，基于健康历史和从头到脚或一般***检查，好的身体评估结果需要一个有组织的、***的方法，对用户的直接反应和实际或潜在的问题的探索发现并研究，因此，如何保证前期用户生理数据的采集量足够的多并且足够全面，如何保证数据处理的合理性减小误差，是进行人体健康评估需要十分关注的问题。
发明内容
本发明的目的是针对现有技术中的不足，提供一种基于睡眠大数据的人体健康评估方法及评估***。该***中的健康大数据研究是指长期地、稳定地、持续增长地获得人体生命特征数据，采用先进的数据挖掘技术清晰明确地描述人类个体的身体状态并预测其健康的未来。通过获取安装于床上的传感器数据，进行人工智能学习模型数据训练，让模型自动学习出与 人体健康相关的特征，从而节省人工挑选和构造特征的时间成本，达到精准预测疾病的目的，进一步通过训练分类器模型可以在健康评估报告中直接获知患病类型。
为实现上述目的，本发明采取的技术方案是：一种基于睡眠大数据的人体健康评估方法，包括以下步骤：
通过安装在床上的传感器，获取人体睡眠时的各项生理数据，并将这些睡眠数据永久地存储于云端服务器；
将上述存储于云端服务器中的睡眠数据进行数据预处理筛除缺失和错误的数据，将正确的数据放入人工智能学习模型中进行训练，使人工智能学习模型学习到患病特征，计算得到生理评价指标；
利用上述人工智能学习模型学习到的数据的患病特征，并以此训练分类器模型，使分类器模型可以识别出不同数据对应的患病类型；
根据上述数据训练得到的人体生理评价指标和分类器模型生成人体健康状态分析报告。
进一步地在某些实施例中，上述方法中经过预处理之后正确的数据分为有患病标签的睡眠数据和无患病标签的睡眠数据，所述有患病标签的睡眠数据来自于已知患病类型的人体，所述无患病标签的睡眠数据来自于未知患病状态的人体。
进一步地在某一实施例中，上述正确的数据放入人工智能学习模型中进行训练的过程包括：
将预处理后无患病标签的睡眠数据送入自编码网络中计算得到无监督网络损失；
将预处理后有患病标签的睡眠数据送入与上述同一参数的自编码网络中，得到中间层降维输出数据并计算出有监督网络损失；
迭代训练上述两个步骤，当得到的无监督网络损失与有监督网络损失之和不再变化或者达到最大迭代步数时，停止迭代，此时得到的睡眠数据即可作为生理评价指标。
进一步地在某一实施例中，上述有患病标签的睡眠数据在自编码网络中得到的中间层降维输出数据，输入至分类器，训练出一个分类器模型。
进一步地，所述生理评价指标用于评估人体是否为患病状态，所述分类器模型用于分析人体患病类型。
为实现上述目的，本发明采取的技术方案进一步是：基于睡眠大数据的人体健康评估***，所述***包括：
睡眠数据获取单元，用于获取人体睡眠时的各项生理数据，所述数据来自于安装于床上的传感器；
睡眠数据存储单元，包括云端服务器，用于存储睡眠数据获取单元采集到的所有睡眠数 据；
数据训练单元，用于对睡眠数据进行预处理，并通过人工智能学习模型对数据进行训练得到生理评价指标；
分类器模型训练单元，用于对人工智能学习模型训练后的睡眠数据进行分类器模型训练；
报告生成单元，用于根据数据训练得到的人体生理评价指标和分类器模型生成人体健康状态分析报告。
进一步地在某一具体实施例中，所述数据训练单元包括：
数据预处理，用于筛除云端服务器中存储的缺失和错误的睡眠数据；
自编码器，用于创建自编码网络，通过计算网络损失对数据进行迭代训练。
为实现上述目的，本发明采取的技术方案进一步是：一种终端设备，所述设备包括上述的基于睡眠大数据的人体健康评估***。所述设备还包括用于控制上述报告生成单元生成报告的操作机构。
本发明优点在于：
1、本发明***与安装在床上传感器实现数据传输，床的使用寿命长久且用户在睡眠状态下的各项生理数据相对稳定，从而保证了数据获取的持久性和稳定性，另一方面用床上安装的传感器获取的海量睡眠数据，为健康评估提供了优质的数据基础。
2、本发明***的云端服务器用于存储用户日常睡眠数据，保证了睡眠数据的稳定性和安全性，同时也为健康评估提供了大量的数据基础，提高评估结果的准确性。
3、本发明的方法通过获取大量人体睡眠时的生理数据，进行人工智能学习模型训练，人在睡眠状态下主动意识非常薄弱，此时得到的生理数据能够真实的反映人体的健康状态；通过搭建适用的人工智能学习模型，让模型自动学习出与人体健康相关的特征，从而节省人工挑选和构造特征的时间成本，达到精准预测疾病的目的。
附图说明
为能更清楚理解本发明的目的、特点和优点，以下将结合附图对本发明的较佳实施例进行详细描述，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。其中：
图1为本发明实施例的基于睡眠大数据的人体健康评估方法的流程示意图；
图2为本发明实施例的人工智能学习模型进行数据训练的流程示意图；
图3为本发明实施例的分类器模型训练的流程示意图；
图4为本发明实施例的基于睡眠大数据的人体健康评估***的框架图。
具体实施方式
为了使本领域的人员更好地理解本发明的技术方案，下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清查、完整地描述，显然，所描述的实施例仅仅是本发明一部分的实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都应当属于本发明保护的范围。
本发明实施例提供一种基于睡眠大数据的人体健康评估方法，用以提高健康评估结果的准确性实现精准预测疾病的目的，适用于安装有传感器的床包括但不仅仅限于：电动床、智能床或者铺设有感应床垫的床等等。图1示出了本发明实施例中基于睡眠大数据的人体健康评估方法的流程图。如附图1所示，本发明实施例中，基于睡眠大数据的人体健康评估方法可以包括：
步骤101.通过安装在床上的传感器，获取人体睡眠时的各项生理数据，并将这些睡眠数据永久地存储于云端服务器。
由图1所示流程可以得知，本发明实施例中基于睡眠大数据的人体健康评估方法是借助人工智能学习模型进行数据训练的方法对存储于云端服务器中大量的睡眠数据进行处理、分析和训练。该方法通过云端服务器与安装于床上的传感器实现数据传输，对于包含有电动控制***的电动床或智能床，云端服务器也可以与控制***的数据库进行通信连接，实时稳定的获取用户的睡眠数据，为健康评估提供优质的数据基础。云端服务器存储的人体睡眠时的各项生理数据类型与安装在床上的传感器类型相关联，实际应用中可以根据健康评估的侧重点选择特定的传感器获取目标数据，例如本实施例中获取到的生理数据包括：心率、呼吸率、翻身、微动、打鼾和离床信息。
步骤102.将上述存储于云端服务器中的睡眠数据进行数据预处理筛除缺失和错误的数据，将正确的数据放入人工智能学习模型中进行训练，使人工智能学习模型学习到患病特征，计算得到生理评价指标。
云端服务器中存储的原始人体睡眠数据为实时获取且没有经过任何处理的，其中难免会含有大量缺失和错误的数据，因此在进行数据分析之前必须经过数据预处理，筛选出正确的睡眠数据才能放入人工智能学习模型中进行训练，不然会导致整个模型错乱，从而无法得到正确的预测结果。数据预处理在筛除噪音数据的同时根据数据来源将睡眠数据分为有患病标签的睡眠数据和无患病标签的睡眠数据，其中有患病标签的睡眠数据是指采集于已知患病类型的人体生理数据中；无患病标签的睡眠数据采集于未知患病状态的人体生理数据中。预处 理后的睡眠数据将被送入人工智能学习模型训练后可用于评估人体的健康状态。具体的实施时，在采集有患病标签的睡眠数据用于人工智能学习模型训练和分类器模型训练时，会选择一个已经准确得知是患病状态，并且知道该疾病类型的个体，对其睡眠数据采集并输入人工智能学习模型中，用来让模型识别人体的状态是正常还是患病以及患病类型。在采集无患病标签的睡眠数据用于人工智能学习模型训练时，会选择一个不能确定是否患病的个体，对其睡眠数据采集并输入人工智能学习模型汇总，用来让模型学习到人体睡眠数据所具有的共性。
步骤103.利用上述人工智能学习模型学习到的数据的患病特征训练分类器模型，使分类器模型可以识别出不同数据对应的患病类型。
在本申请中，经过人工智能学习模型训练的数据将作为生理评价指标，可以用来识别个体的身体状态为患病状态还是正常状态。进一步地，将有患病标签的睡眠数据通过在自编码网络的中间层降维输出后，作为SVM分类器的输入，用来训练分类器模型，使分类器模型可以识别出不同数据对应的患病类型。经过步骤102使人工智能学习模型从原始数据中学习到和患病相关的一些特征，再经过步骤103进一步训练使分类器模型可以识别出产生该睡眠数据的个体所患疾病的具体类型。
步骤104.根据上述数据训练得到的人体生理评价指标和分类器模型生成人体健康状态分析报告。
下面结合附图1-3介绍本发明的具体实施方式。
首先将存储在云端服务器中的睡眠数据进行数据预处理，筛除去缺失和错误的数据，剩下的正确的睡眠数据根据数据来源分为有患病标签的睡眠数据和无患病标签的睡眠数据，将这些数据放入人工智能学习模型中进行训练。
附图2示出了本发明实施例的人工智能学习模型进行数据训练的流程示意图，如图2所示，经过预处理后的正确的数据放入人工智能学习模型中进行训练的过程包括：
(1)将预处理后无患病标签的睡眠数据x送入自编码网络的深度自编码器中，计算得到无监督网络损失loss1 k。本申请中选用最优的计算方式为：将睡眠数据x送入深度自编码器中得到输入和输出的均方差值，并将该值赋予无监督网络损失loss1 k。
(2)将预处理后有患病标签的睡眠数据y送入与上述同一参数的自编码网络中，得到中间层降维输出数据并计算出有监督网络损失loss2 k。本申请中选用最优的计算有监督网络损失loss2 k的方式为：根据患病标签将得到的中间层降维输出数据分成不同的类别，找出每个类别的中心点，并计算同类数据到同类中心点的距离，所得数据之和赋予有监督网络损失loss2 k。
(3)将上述(1)和(2)得到的loss1 k和loss2 k求和即得网络总损失loss k,即 loss k＝loss1 k+loss2 k，利用反向传播算法调整整个网络结构参数。然后采用迭代训练方式，将x和y再次送入自编码网络结构中，分别重复步骤(1)和步骤(2)再次计算网络结构参数调整之后的无监督网络损失loss1 k+1和有监督网络损失loss2 k+1，并计算得到本轮网络总损失loss k+1＝loss1 k+1+loss2 k+1。
(4)比较调整网络结构参数后的损失值并再次调整网络结构参数，直到网络结构的loss k+1＝loss k或者达到最大迭代次数时停止训练。最大迭代次数是指，在人工智能学习模型的训练过程中，根据需要设定最大迭代次数M，当迭代次数为M(k＝M)时，即使网络总损失仍然发生变化也不会继续进行迭代训练。
至此人工智能学习模型训练完成，训练得到的神经网络损失和自编码网络损失将作为生理评价指标，用于评估睡眠数据对应的个体是否为患病状态或者为正常的健康状态。
上述有患病标签的睡眠数据y送入自编码网络中时所得到的中间层降维输出数据，进一步将被作为SVM分类器的输入，用来训练分类器模型，使分类器模型可以识别出不同数据对应的患病类型。训练得到的分类器模型可以识别出产生该睡眠数据的个体所患疾病的具体类型。
经过上述训练得到的生理评价指标和分类器模型，既可以用于生成睡眠数据对应的个体健康状态分析报告。具体实施时，从云端服务器中选出需要进行健康评估的个体的睡眠数据，将这些数据放入已经训练完成的人工智能学习模型中，通过该模型确定该个体的患病状态，如得到结果为患病状态，则进一步的可以得到该个体所患病类型。最终可以借助终端设备输出或打印出包含上述评估结果的人体健康评估报告，实现利用睡眠数据预测疾病的目的。
图4为本发明实施例的基于睡眠大数据的人体健康评估***的框架图，如附图4所示，基于睡眠大数据的人体健康评估***包括：
睡眠数据获取单元，用于获取人体睡眠时的各项生理数据，所述数据来自于安装于床上的传感器；
睡眠数据存储单元，包括云端服务器，用于存储睡眠数据获取单元采集到的所有睡眠数据；
数据训练单元，用于对睡眠数据进行预处理，并通过人工智能学习模型对数据进行训练得到生理评价指标。数据训练单元包括：数据预处理，用于筛除云端服务器中存储的缺失和错误的睡眠数据；自编码器，用于创建自编码网络，通过计算网络损失对数据进行迭代训练。
分类器模型训练单元，用于对人工智能学习模型训练后的睡眠数据进行分类器模型训练；
报告生成单元，用于根据数据训练得到的人体生理评价指标和分类器模型生成人体健康状态分析报告。
需要说明的是，本发明***可以设置于电动床的控制***中或者装设于人机交互操作的终端设备中，设备还包括用于控制报告生成单元生成报告的操作机构，以便于用户可以按照需求对评估***进行自定义设置。本说明书实施例中所述的终端设备可以包括用户设备、智能手机、电脑、移动互联网设备或穿戴式智能设备等互联网设备，本发明实施例不作限定。
本领域内的技术人员应明白，本发明的实施例可提供为方法、***或计算机程序产品。因此，本发明可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。
以上所述仅是本发明的优选实施方式，应当指出，对于本技术领域的普通技术人员，在不脱离本发明方法的前提下，还可以做出若干改进和补充，这些改进和补充也应视为本发明的保护范围。

Claims (10)

  1. 一种基于睡眠大数据的人体健康评估方法，其特征在于，包括：
    通过安装在床上的传感器，获取人体睡眠时的各项生理数据，并将这些睡眠数据永久地存储于云端服务器；
    将上述存储于云端服务器中的睡眠数据进行数据预处理筛除缺失和错误的数据，将正确的数据放入人工智能学习模型中进行训练，使人工智能学习模型学习到患病特征，计算得到生理评价指标；
    利用上述人工智能学习模型学习到数据的患病特征，训练分类器模型，使分类器模型可以识别出不同数据对应的患病类型；
    根据上述数据训练得到的人体生理评价指标和分类器模型生成人体健康状态分析报告。
  2. 根据权利要求1所述的基于睡眠大数据的人体健康评估方法，其特征在于，人体睡眠时的各项生理数据包括：心率、呼吸率、翻身、微动、打鼾和离床信息。
  3. 根据权利要求1所述的基于睡眠大数据的人体健康评估方法，其特征在于，经过预处理之后正确的数据分为有患病标签的睡眠数据和无患病标签的睡眠数据，所述有患病标签的睡眠数据来自于已知患病类型的人体，所述无患病标签的睡眠数据来自于未知患病状态的人体。
  4. 根据权利要求3所述的基于睡眠大数据的人体健康评估方法，其特征在于，所述正确的数据放入人工智能学习模型中进行训练的过程包括：
    将预处理后无患病标签的睡眠数据送入自编码网络中计算得到无监督网络损失；
    将预处理后有患病标签的睡眠数据送入与上述同一参数的自编码网络中，得到中间层降维输出数据并计算出有监督网络损失；
    迭代训练上述两个步骤，当得到的无监督网络损失与有监督网络损失之和不再变化或者达到最大迭代步数时，停止迭代，此时得到的睡眠数据即可作为生理评价指标。
  5. 根据权利要求4所述的基于睡眠大数据的人体健康评估方法，其特征在于，将所述有患病标签的睡眠数据在自编码网络中得到的中间层降维输出数据，输入至分类器，训练出一个分类器模型。
  6. 根据权利要求5所述的基于睡眠大数据的人体健康评估方法，其特征在于，所述生理评价指标用于评估人体是否为患病状态，所述分类器模型用于分析人体患病类型。
  7. 基于睡眠大数据的人体健康评估***，其特征在于，包括：
    睡眠数据获取单元，用于获取人体睡眠时的各项生理数据，所述数据来自于安装于床上的传感器；
    睡眠数据存储单元，包括云端服务器，用于存储睡眠数据获取单元采集到的所有睡眠数 据；
    数据训练单元，用于对睡眠数据进行预处理，并通过人工智能学习模型对数据进行训练得到生理评价指标；
    分类器模型训练单元，用于对人工智能学习模型训练后的睡眠数据进行分类器模型训练；
    报告生成单元，用于根据数据训练得到的人体生理评价指标和分类器模型生成人体健康状态分析报告。
  8. 根据权利要求7所述的基于睡眠大数据的人体健康评估***，其特征在于，所述数据训练单元包括：
    数据预处理，用于筛除云端服务器中存储的缺失和错误的睡眠数据；
    自编码器，用于创建自编码网络，通过计算网络损失对数据进行迭代训练。
  9. 一种终端设备，其特征在于，所述设备包括权利要求7或8所述的基于睡眠大数据的人体健康评估***。
  10. 根据权利要求9所述的一种终端设备，其特征在于，所述设备还包括用于控制所述权利要求7中报告生成单元生成报告的操作机构。
PCT/CN2018/100955 2018-07-13 2018-08-17 基于睡眠大数据的人体健康评估方法及评估*** WO2020010668A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
RU2020129796A RU2757048C1 (ru) 2018-07-13 2018-08-17 Способ и система оценки здоровья тела человека на основе данных большого объема о сне
AU2018432124A AU2018432124A1 (en) 2018-07-13 2018-08-17 Human body health assessment method and system based on sleep big data
EP18926215.7A EP3783619A4 (en) 2018-07-13 2018-08-17 METHOD AND SYSTEM FOR ASSESSING THE HEALTH OF A HUMAN BODY BASED ON LARGE AMOUNTS OF SLEEP DATA
US17/040,975 US20210225510A1 (en) 2018-07-13 2018-08-17 Human body health assessment method and system based on sleep big data

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810772360.1 2018-07-13
CN201810772360.1A CN109087706B (zh) 2018-07-13 2018-07-13 基于睡眠大数据的人体健康评估方法及评估***

Publications (1)

Publication Number Publication Date
WO2020010668A1 true WO2020010668A1 (zh) 2020-01-16

Family

ID=64837905

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/100955 WO2020010668A1 (zh) 2018-07-13 2018-08-17 基于睡眠大数据的人体健康评估方法及评估***

Country Status (6)

Country Link
US (1) US20210225510A1 (zh)
EP (1) EP3783619A4 (zh)
CN (1) CN109087706B (zh)
AU (1) AU2018432124A1 (zh)
RU (1) RU2757048C1 (zh)
WO (1) WO2020010668A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111696679A (zh) * 2020-07-14 2020-09-22 章越新 一种基于量子Petri网的人体健康状态分析方案
CN116269242A (zh) * 2023-05-17 2023-06-23 广州培生智能科技有限公司 一种基于互联网的老年人健康状况实时监控***

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109767836A (zh) * 2018-12-29 2019-05-17 上海亲看慧智能科技有限公司 一种医学诊断人工智能***、装置及其自我学习方法
CN109661039B (zh) * 2019-01-15 2020-07-21 北京泰德东腾通信技术有限公司 5g会话建立及释放的协议一致性测试方法及***
CN109840932A (zh) * 2019-01-23 2019-06-04 平安科技（深圳）有限公司 心脑血管疾病展示方法、装置、设备及存储介质
CN109770881A (zh) * 2019-01-28 2019-05-21 深圳市星裳尔电子商务有限公司 一种基于血压的健康监测方法及血压测量装置
JP7326926B2 (ja) * 2019-06-27 2023-08-16 トヨタ自動車株式会社 学習装置、リハビリ支援システム、方法、プログラム、及び学習済みモデル
CN111009328A (zh) * 2019-12-25 2020-04-14 新绎健康科技有限公司 一种区域人口健康状况评估方法及装置
CN111785374A (zh) * 2020-06-15 2020-10-16 山东省玖玖医养健康产业有限公司 一种基于大数据的健康状况分析预测方法及***
US10931643B1 (en) * 2020-07-27 2021-02-23 Kpn Innovations, Llc. Methods and systems of telemedicine diagnostics through remote sensing
WO2022208338A1 (en) * 2021-03-30 2022-10-06 Jio Platforms Limited System and method of data ingestion and processing framework
CN117235200B (zh) * 2023-09-12 2024-05-10 杭州湘云信息技术有限公司 基于ai技术的数据集成方法、装置、计算机设备及存储介质
CN117992803B (zh) * 2024-04-03 2024-06-11 江西千策信息工程有限公司 一种基于钢结构健康的多尺度结合监测方法及***

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN204971278U (zh) * 2015-08-24 2016-01-20 华南理工大学 基于健康服务机器人的老年痴呆症监护***
WO2016201499A1 (en) * 2015-06-15 2016-12-22 Medibio Limited Method and system for assessing mental state
CN106618611A (zh) * 2017-03-06 2017-05-10 兰州大学 基于睡眠多通道生理信号的抑郁症辅助诊断方法和***
CN108091391A (zh) * 2017-12-27 2018-05-29 深圳和而泰数据资源与云技术有限公司 病症评估方法、终端设备及计算机可读介质

Family Cites Families (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2574275A3 (en) * 2004-03-22 2013-06-26 BodyMedia, Inc. Non-Invasive Temperature Monitoring Device
US7733224B2 (en) * 2006-06-30 2010-06-08 Bao Tran Mesh network personal emergency response appliance
US7558622B2 (en) * 2006-05-24 2009-07-07 Bao Tran Mesh network stroke monitoring appliance
WO2011009085A2 (en) * 2009-07-17 2011-01-20 Oregon Health & Science University Method and apparatus for assessment of sleep disorders
US9955795B2 (en) * 2014-06-05 2018-05-01 Matthew W. Krenik Automated bed and method of operation thereof
US20180014784A1 (en) * 2015-01-30 2018-01-18 New York University System and method for electrophysiological monitoring
WO2016142360A1 (en) * 2015-03-09 2016-09-15 Koninklijke Philips N.V. Wearable device obtaining audio data for diagnosis
US10105092B2 (en) * 2015-11-16 2018-10-23 Eight Sleep Inc. Detecting sleeping disorders
CN105997053B (zh) * 2016-06-25 2019-04-12 浙江和也健康科技有限公司 一种基于智能寝具的健康管理***
CN110072432B (zh) * 2016-07-29 2022-12-09 布莱特有限公司 使用数据分析和学习技术以改善个人睡眠条件的自适应性睡眠***
CN106156530A (zh) * 2016-08-03 2016-11-23 北京好运到信息科技有限公司 基于栈式自编码器的体检数据分析方法及装置
CN107103182A (zh) * 2017-03-28 2017-08-29 南京医科大学 一种基于深度学习算法的心脏性疾病风险预警***及方法
CN107595243B (zh) * 2017-07-28 2021-08-17 深圳和而泰智能控制股份有限公司 一种病症评估方法及终端设备
JP7437575B2 (ja) * 2018-01-05 2024-02-26 スリープ ナンバー コーポレイション 生理学的イベント検知特徴部を備えたベッド
US11364361B2 (en) * 2018-04-20 2022-06-21 Neuroenhancement Lab, LLC System and method for inducing sleep by transplanting mental states
US10890968B2 (en) * 2018-05-07 2021-01-12 Apple Inc. Electronic device with foveated display and gaze prediction

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016201499A1 (en) * 2015-06-15 2016-12-22 Medibio Limited Method and system for assessing mental state
CN204971278U (zh) * 2015-08-24 2016-01-20 华南理工大学 基于健康服务机器人的老年痴呆症监护***
CN106618611A (zh) * 2017-03-06 2017-05-10 兰州大学 基于睡眠多通道生理信号的抑郁症辅助诊断方法和***
CN108091391A (zh) * 2017-12-27 2018-05-29 深圳和而泰数据资源与云技术有限公司 病症评估方法、终端设备及计算机可读介质

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3783619A4 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111696679A (zh) * 2020-07-14 2020-09-22 章越新 一种基于量子Petri网的人体健康状态分析方案
CN116269242A (zh) * 2023-05-17 2023-06-23 广州培生智能科技有限公司 一种基于互联网的老年人健康状况实时监控***
CN116269242B (zh) * 2023-05-17 2023-07-18 广州培生智能科技有限公司 一种基于互联网的老年人健康状况实时监控***

Also Published As

Publication number Publication date
US20210225510A1 (en) 2021-07-22
CN109087706B (zh) 2022-12-02
EP3783619A1 (en) 2021-02-24
EP3783619A4 (en) 2022-01-26
CN109087706A (zh) 2018-12-25
RU2757048C1 (ru) 2021-10-11
AU2018432124A1 (en) 2020-10-22

Similar Documents

Publication Publication Date Title
WO2020010668A1 (zh) 基于睡眠大数据的人体健康评估方法及评估***
CN104287706B (zh) 一种人体健康状态实时监测及诊疗推荐***
US20210035067A1 (en) Method to increase efficiency, coverage, and quality of direct primary care
CN109119130A (zh) 一种基于云计算的大数据健康管理***及方法
CN108778097A (zh) 用于评估心力衰竭的装置和方法
CN108366763A (zh) 用于评估精神状态的方法和***
US11529105B2 (en) Digital twin updating
CN110584601B (zh) 一种老人认知功能监测和评估***
CN109065162A (zh) 一种综合性智能化诊断***
CN108814565A (zh) 一种基于多传感器信息融合和深度学习的智能中医健康检测梳妆台
Elbayoudi et al. The human behaviour indicator: A measure of behavioural evolution
CN108652587A (zh) 一种认知功能障碍预防监测装置
Gan et al. Human-computer interaction based interface design of intelligent health detection using PCANet and multi-sensor information fusion
CN117064388A (zh) 基于情绪识别实现精神障碍评估分析的***
CN116434979A (zh) 生理状态云监测方法、监测***及储存介质
Fazeli et al. Passive monitoring of physiological precursors of stress leveraging smartwatch data
CN115472312A (zh) 一种基于医防融合服务的医疗健康管理***及方法
WO2022272057A1 (en) Devices, systems, and methods for mental health assessment
CN114388089A (zh) 一种基于人工智能的个人健康管理方法及***
Lim et al. Developing a mobile wellness management system for healthy lifestyle by analyzing daily living activities
CN112735590A (zh) 一种基于中医大数据分析的腰椎突症疗效评价专家***
Song et al. [Retracted] Health Promotion Effects of Sports Training Based on HMM Theory and Big Data
Lin et al. Applications of non‐invasive sensor devices to personalised health care
KR102478102B1 (ko) 불면증 치료방법
KR102462574B1 (ko) 만성질환자의 순응도 및 상태 예측을 위한 인공지능학습 모델의 학습 방법 및 그 시스템

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18926215

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2018432124

Country of ref document: AU

Date of ref document: 20180817

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 18926215.7

Country of ref document: EP

ENP Entry into the national phase

Ref document number: 2018926215

Country of ref document: EP

Effective date: 20201117

NENP Non-entry into the national phase

Ref country code: DE