Old people health management detection method and system

Abstract

The invention provides an old people health management detection method and system, and the method comprises the steps: obtaining behavior task data and physiological data of old people through a scale or a non-invasive device, extracting data features, carrying out the fusion, building a cognitive state evaluation model set through combining an integrated learning model, a deep learning model and a statistical learning model, and carrying out the detection of the health management of old people. Training a cognitive state evaluation model set by adopting a differentiated learning strategy, constructing a dynamic evaluation system based on cognitive competence scores, determining evaluation frequency, determining evaluation items by adopting a correlation priority strategy of data features, establishing a nonlinear mapping rule, setting training reference parameters according to the cognitive competence scores, and evaluating the cognitive competence scores according to the training reference parameters. And making a personalized cognitive training scheme based on the dynamic evaluation system and the individual data characteristics of the elderly. The method is high in individuation degree and simple to operate, has continuous and comprehensive data, and can capture dynamic changes of cognitive competence, know differences of old people in real time and provide a timely and effective feedback mechanism.

Description

Technical Field

[0001] This application relates to the technical field of cognitive testing for the elderly, and specifically relates to a health management detection method and system for the elderly. Background Art

[0002] The core of cognitive health management lies in timely and accurately evaluating the cognitive status of the elderly, and providing personalized intervention measures according to the evaluation results to delay the process of cognitive decline and improve the quality of life of the elderly.

[0003] Traditional methods for assessing elderly cognition mainly include paper-and-pencil tests, neuropsychological assessments, etc. In recent years, with the rapid development of information technology, technologies based on electronic devices and sensors have gradually been applied to the field of cognitive health management. Existing applications for monitoring and managing elderly cognition include neuroscale assessments, eye movement technology assessments, and professional equipment assessments; Neuroscale assessments utilize data on the cognitive functions of the elderly, such as attention and memory, but the content of the flow meter is long and highly subjective, and there are many interfering factors in actual applications; Eye movement technology assessments combine the paradigms of saccade, antisaccade, and random saccade with eye movement technology to collect the fixation paths and regions of the elderly when observing specific visual stimuli, but this method cannot be mutually verified with other assessment methods at the same time; Professional equipment assessments are achieved through specialized medical equipment and data analysis platforms for data transmission, collection, and management, but the cost of this method is high and it cannot be used as a large-scale popular cognitive screening tool. In addition, the above methods also have many limitations, specifically including: Lack of personalized assessment and training programs: Although some applications use different assessment methods, most use a unified standard to evaluate the cognitive abilities of the elderly, failing to fully consider individual differences, resulting in subsequent training programs lacking pertinence and being unable to meet the specific needs of different elderly people.

[0004] The assessment process is cumbersome and inconvenient: Traditional cognitive assessments usually require professional personnel for on-site guidance and operation, involving multiple complex test links, which may pose problems such as difficult operation and low participation for the elderly, and are also not convenient for long-term and frequent monitoring.

[0005] Incomplete data collection and analysis: Existing assessment methods often can only obtain limited cognitive data, such as the results of simple questionnaire tests, and cannot comprehensively and dynamically monitor the cognitive performance of the elderly in daily life, making it difficult to detect subtle changes in cognitive abilities in a timely manner.

[0006] Lack of real-time feedback and guidance mechanism: When the elderly are undergoing assessment or training, they often cannot understand their performance and progress in real time, and cannot adjust their behaviors and training methods in a timely manner according to the feedback, affecting the effect of health management.

[0007] This application aims to solve the above technical problems and provides a health management detection method and system for the elderly that can achieve personalized assessment and training, is easy to operate, comprehensively collects data, and provides real-time feedback. Summary of the Invention

[0008] In view of one or more technical deficiencies in the above prior art, the following technical solutions are proposed in this application.

[0009] Based on the first aspect of this application, a health management detection method for the elderly is proposed, including: S1: Collect the behavioral task data and physiological data of the elderly through a scale or non-invasive device, and transmit the behavioral task data and the physiological data to the cloud; S2: Extract data features from the behavioral task data and physiological data and fuse them. The data features include task completion status, reaction time, and visual behavior patterns; S3: Weight the ensemble learning model, deep learning model, and statistical learning model according to certain weights to construct a cognitive state assessment model set, train the cognitive state assessment model set using a differential learning strategy, analyze the importance of the data features through a decision tree model, quantify the correlation strength between the cognitive level of the elderly and the data features through a regression model, and use the trained cognitive state assessment model set to score the behavioral task data and physiological data to obtain a cognitive ability score; S4: Build a dynamic assessment system based on the cognitive ability score, determine the assessment frequency through the cognitive ability score, determine the assessment items using the correlation priority strategy of the data features, establish a non-linear mapping rule, enable a simplified assessment process when the cognitive ability score is lower than a preset threshold, and add high-order executive function test items when the cognitive ability score is higher than the preset threshold to form a stepped assessment plan; S5: Set training benchmark parameters according to the cognitive ability score, real-time track the behavioral task data and physiological data of the elderly, formulate a personalized cognitive training plan based on the dynamic assessment system and the individual data features of the elderly, compare the training performance of the elderly, and provide feedback through personalized guidance suggestions.

[0010] Furthermore, the behavioral task data includes graphic recognition data, memory task data, scale data, interface operation trajectory data, task completion time data, and attention test data; The physiological data includes blood oxygen saturation data, electroencephalogram signal data, heart rate data, and eye movement data.

[0011] Collect physiological data through non-invasive devices and collect behavioral task data through a simple and easy-to-use visual interface design, which can not only ensure the continuity and comprehensiveness of the data, but also ensure the simplicity of operation for the elderly.

[0012] Furthermore, step S3 also includes using a support vector machine (SVM) to process the electroencephalogram signal data and blood oxygen saturation data, dealing with the non-linearly separable problem through kernel function transformation, and analyzing the correlation between eye movement data and scale data.

[0013] Furthermore, step S3 also includes using a Naive Bayes classifier to make conditional independence assumptions on the task completion time data and interface operation trajectory data and perform rapid inference.

[0014] Furthermore, step S3 also includes calculating the Gain value for the eye movement data and scale data, sorting the importance of data features, and screening key indicators.

[0015] Furthermore, step S3 also includes using a time series algorithm to capture the long-range dependence relationship of heart rate data through a memory gating mechanism and reduce the mean absolute error.

[0016] Adopting a differential learning strategy to train the cognitive state evaluation model set can ensure the accuracy and robustness of the model output results.

[0017] Furthermore, step S5 also includes adjusting the personalized cognitive training plan through reinforcement learning and state-action-reward mechanism.

[0018] Formulating and timely adjusting a personalized cognitive training plan for the elderly can help the elderly timely understand their training progress and improve their training enthusiasm, and improve the accuracy and reliability of cognitive evaluation.

[0019] Based on the second aspect of the present application, an elderly health management detection system is also proposed, including: Acquisition module: Collect the behavioral task data and physiological data of the elderly through a scale or non-invasive device, and transmit the behavioral task data and the physiological data to the cloud; Fusion module: Extract data features from the behavioral task data and physiological data and perform fusion, and the data features include task completion situation, reaction time, and visual behavior pattern; Analysis module: Weight the integrated learning model, deep learning model, and statistical learning model according to certain weights to construct a set of cognitive state evaluation models. Adopt a differential learning strategy to train the set of cognitive state evaluation models. Analyze the importance of the data features through a decision tree model, quantify the correlation strength between the cognitive level of the elderly and the data features through a regression model, and use the trained set of cognitive state evaluation models to score the behavior task data and physiological data to obtain a cognitive ability score. Evaluation module: Based on the cognitive ability score, construct a dynamic evaluation system. Determine the evaluation frequency through the cognitive ability score, adopt the correlation degree priority strategy of the data features to determine the evaluation items, establish a non-linear mapping rule. When the cognitive ability score is lower than a preset threshold, enable a simplified evaluation process. When the cognitive ability score is higher than the preset threshold, add high-order executive function test items to form a stepped evaluation plan. Feedback module: Set training benchmark parameters according to the cognitive ability score, real-time track the behavior task data and physiological data of the elderly, formulate a personalized cognitive training plan based on the dynamic evaluation system and the individual data features of the elderly, compare the training performance of the elderly and give feedback through personalized guidance suggestions.

[0020] Based on the third aspect of the present application, a computer program product is also proposed, which has one or more computer programs. When the computer program is executed by a computer processor, the method described in any one of the above is implemented.

[0021] The technical effect of the present application is as follows: The present application combines and mutually verifies multiple evaluation methods, increases the reliability of the cognitive evaluation results without increasing the evaluation complexity, solves the problem of the separation of behavior task data and physiological data in the traditional cognitive ability evaluation of the elderly, and the problem of the single traditional cognitive ability evaluation method of the elderly, avoids a one-size-fits-all evaluation method and training method, improves the reliability of the cognitive ability evaluation, designs different personalized evaluation training plans for the individual differences of the elderly, accurately adapts to the evaluation difficulty, and effectively improves the effect of health management. Brief Description of the Drawings

[0022] By reading the detailed description of the non-restrictive embodiments with reference to the following drawings, other features, purposes, and advantages of the present application will become more obvious.

[0023] Figure 1 is the overall flowchart of a detection method for elderly health management provided by an embodiment of the present application; Figure 2 is the overall framework diagram of a detection system for elderly health management provided by an embodiment of the present application; Figure 3 It is a specific operation diagram of an elderly health management detection system provided according to an embodiment of the present application; Figure 4 It is a schematic structural diagram of a computer system of an electronic device suitable for implementing the embodiments of the present application. Detailed implementation manners

[0024] The present application will be further described in detail below with reference to the drawings and embodiments. It can be understood that the specific embodiments described herein are only used to explain the related invention, rather than limiting the invention. Additionally, it should be noted that for the convenience of description, only parts related to the invention are shown in the drawings.

[0025] It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The present application will be described in detail below with reference to the drawings and embodiments.

[0026] Below refer to Figure 1 , Figure 1 which shows an elderly health management detection method, including: S1: Collect the behavioral task data and physiological data of the elderly through a scale or non-invasive device, and transmit the behavioral task data and the physiological data to the cloud; S2: Extract data features from the behavioral task data and physiological data and perform fusion. The data features include task completion, reaction time, and visual behavior patterns; S3: Weight an ensemble learning model, a deep learning model, and a statistical learning model according to certain weights and construct a cognitive state evaluation model set. Use a differential learning strategy to train the cognitive state evaluation model set. Analyze the importance of the data features through a decision tree model, quantify the correlation strength between the cognitive level of the elderly and the data features through a regression model, and use the trained cognitive state evaluation model set to score the behavioral task data and physiological data to obtain a cognitive ability score; S4: Build a dynamic evaluation system based on the cognitive ability score. Determine the evaluation frequency through the cognitive ability score, determine the evaluation items using the correlation priority strategy of the data features, establish a non-linear mapping rule, enable a simplified evaluation process when the cognitive ability score is lower than a preset threshold, and add high-order executive function test items when the cognitive ability score is higher than the preset threshold to form a stepped evaluation plan; S5: Set training benchmark parameters according to the cognitive ability score, real-time track the behavioral task data and physiological data of the elderly, formulate a personalized cognitive training plan based on the dynamic evaluation system and the individual data features of the elderly, compare the training performance of the elderly, and provide feedback through personalized guidance suggestions.

[0027] It should be noted that the behavioral task data includes graphic recognition data, memory task data, scale data, interface operation trajectory data, task completion time data, and attention test data; The physiological data includes blood oxygen saturation data, electroencephalogram (EEG) signal data, heart rate data, and eye movement data.

[0028] It should be noted that step S3 further includes using a support vector machine (SVM) to process the EEG signal data and blood oxygen saturation data, dealing with the non-linearly separable problem through kernel function transformation, and analyzing the correlation between eye movement data and scale data.

[0029] It should be noted that step S3 further includes using a naive Bayes classifier to make a conditional independence assumption for the task completion time data and interface operation trajectory data and perform rapid inference.

[0030] It should be noted that step S3 further includes calculating the Gain value for the eye movement data and scale data, sorting the importance of data features, and screening key indicators.

[0031] It should be noted that step S3 further includes using a time series algorithm to capture the long-range dependence relationship of heart rate data through a memory gating mechanism and reduce the mean absolute error.

[0032] It should be noted that step S5 further includes adjusting the personalized cognitive training plan through reinforcement learning and a state-action-reward mechanism.

[0033] It should be noted that the test task operation is completed by touching the screen to obtain behavioral task data, and physiological data is collected through non-invasive devices. The non-invasive devices include cameras, scales, and sensors. This application combines physiological data and behavioral task data, improving the accuracy and reliability of cognitive assessment and ensuring the continuity and comprehensiveness of data collection.

[0034] It should be noted that the behavioral task data and the physiological data are transmitted to the cloud through cloud transmission technology, and the data is encrypted and integrity verified to ensure the integrity and confidentiality of the data.

[0035] In a specific embodiment, an eye movement tracking system using infrared light projects infrared light onto the user's eyes to detect changes in the reflected light of the eyes to obtain eye movement data, which has higher accuracy and stability; An EEG sensor is used to measure the electrical activity of the brain and analyze the relationship between specific frequency bands (such as alpha waves, beta waves, etc.) in the EEG signal and the cognitive state.

[0036] In a specific embodiment, in the scenarios of real-time monitoring and remote health management, high-speed wireless communication technologies such as Wi-Fi, Bluetooth, or 5G are used to transmit data to the cloud. 5G has higher bandwidth and lower latency, enabling faster and more stable transmission of a large amount of cognitive data.

[0037] It should be noted that the integrated learning model includes random forest and XGBoost, the deep learning model includes LSTM and graph convolutional network, and the statistical learning model includes support vector machine and Gaussian process model.

[0038] In a specific embodiment, preferably, a basic model group is constructed by weighted aggregation of a random forest model, a bidirectional LSTM model, and a support vector regression model in a ratio of 6:3:1. Then, through stratified five-fold cross-validation, the age and gender distributions of each fold of data are kept consistent, ensuring that the correlation coefficient of the cognitive ability evaluation results output by each model reaches above 0.85.

[0039] In a specific embodiment, the range of the cognitive ability score value is 0 - 100 points. For the moderately cognitively declined group with a cognitive ability score of 40 - 60 points, an evaluation frequency of twice a week is adopted. If the standard deviation of the cognitive ability score value exceeds 8 points for three consecutive times, the dynamic encryption mechanism is triggered and the evaluation frequency is changed to three times a week to ensure timely capture of the cognitive fluctuations of the elderly.

[0040] In a specific embodiment, the correlation weights between the physiological data and behavioral task data of the elderly and the cognitive ability score are calculated through a random forest model, and the three evaluation contents with the highest correlation degree are matched for each user. For example, for an elderly person with a cognitive ability score of 65 points and a brain wave correlation degree of 0.32, the system focuses on deploying the attention persistence test.

[0041] In a specific embodiment, the evaluation of the elderly's cognitive ability includes evaluation items, evaluation frequency, and evaluation difficulty; when the elderly's cognitive ability score is lower than 30 points, a simplified evaluation process is enabled, such as shortening the evaluation duration by 30% or canceling the evaluation time limit; when the elderly's cognitive ability score is higher than 60 points, high-order executive function test items are added.

[0042] It should be noted that the personalized training plan includes training content, training duration, and training difficulty. In the embodiment, the 20% cognitive load redundancy design principle is adopted, that is, the training difficulty is 120% of the current cognitive ability score value. Benchmark parameters are set according to the interval where the cognitive ability score is located. For example, the training duration of the elderly with a cognitive ability score of 50 - 70 points is set to 35 minutes, and closed-loop optimization is achieved by real-time tracking of the training completion degree and physiological data; When the completion rate of three consecutive training tasks is lower than 60% and the decrease in heart rate variability (HRV) is less than 15%, a three-level downgrading mechanism is triggered. The training difficulty is reduced by 10%, the training duration is shortened by 20%, and 25% of the training items are replaced. Conversely, if the improvement in the cognitive ability score of the elderly exceeds 5 points, a stepped reinforcement training is initiated. The training difficulty is increased by 8% per week and a cross-modal training task is added.

[0043] In a specific embodiment, the state-action-reward mechanism obtains a state value based on the cognitive ability score and the HRV trend, adjusts the action difficulty of the training task according to the state value, and gives appropriate rewards according to the task completion improvement rate. The state-action-reward mechanism realizes self-evolution of personalized adjustment, improving the cognitive training efficiency of the elderly by 41%.

[0044] It should be noted that during the training of the elderly, the correct rate and reaction time are displayed in real time through a visual interface, compared with the previous training results, a reward and incentive mechanism is set, and targeted suggestions and guidance are provided according to the training results and cognitive status of the elderly. When the elderly perform poorly in a certain training item, specific training methods and skills are provided.

[0045] It should be noted that this application evaluates the fixation point coordinates and saccade frequency in the eye movement data and the option dwell time and correction times in the scale data, performs time series alignment through the dynamic time warping algorithm, constructs an attention weight matrix to achieve feature fusion, and combines multiple methods of ensemble verification, solving the problem of the separation of behavioral and physiological data in traditional evaluations and the shortcoming of the single traditional cognitive ability evaluation method, improving the reliability of cognitive evaluation.

[0046] It should be noted that for small-sample physiological data and sparse physiological data intermittently collected by wearable devices, a support vector machine model is used, and the support vector machine model shows high robustness in data correlation analysis; For the training difficulty adjustment scenario that requires real-time feedback, a naive Bayes classifier is used; Using an ensemble learning algorithm (such as XGBoost and LightGBM) to process noisy data can stabilize the model AUC value between 0.82 and 0.91; For the scenario of continuous monitoring of physiological indicators and cognitive decline trend prediction tasks, time series algorithms (such as LSTM and TCN) are used, which can reduce the mean absolute error by 23% - 37%; For the dynamic training plan optimization scenario, using a reinforcement learning algorithm (such as PPO and DQN) can achieve self-evolution of personalized strategies, improving the training efficiency by 41%.

[0047] The following refers to Figure 2 ,Figure 2 A health management detection system for the elderly is shown, including an acquisition module a, a fusion module b, an analysis module c, an evaluation module d, and a feedback model e.

[0048] In a specific embodiment, the acquisition module a is configured to: collect the behavioral task data and physiological data of the elderly through a scale or non-invasive device, and transmit the behavioral task data and the physiological data to the cloud.

[0049] In a specific embodiment, the fusion module b is configured to: extract data features from the behavioral task data and physiological data and perform fusion, and the data features include task completion, reaction time, and visual behavior patterns.

[0050] In a specific embodiment, the analysis module c is configured to: weight an ensemble learning model, a deep learning model, and a statistical learning model according to certain weights and construct a cognitive state evaluation model set, train the cognitive state evaluation model set using a differential learning strategy, analyze the importance of the data features through a decision tree model, quantify the correlation strength between the cognitive level of the elderly and the data features through a regression model, and use the trained cognitive state evaluation model set to score the behavioral task data and physiological data to obtain a cognitive ability score.

[0051] In a specific embodiment, the evaluation module d is configured to: construct a dynamic evaluation system based on the cognitive ability score, determine the evaluation frequency through the cognitive ability score, determine the evaluation items using the correlation degree priority strategy of the data features, establish a non-linear mapping rule, enable a simplified evaluation process when the cognitive ability score is lower than a preset threshold, and add high-order executive function test items when the cognitive ability score is higher than the preset threshold to form a stepped evaluation plan.

[0052] In a specific embodiment, the feedback module e is configured to: set training benchmark parameters according to the cognitive ability score, real-time track the behavioral task data and physiological data of the elderly, formulate a personalized cognitive training plan based on the dynamic evaluation system and the individual data features of the elderly, compare the training performance of the elderly and provide feedback through personalized guidance suggestions.

[0053] It should be noted that the data collected in this application is comprehensive and continuous, integrating behavioral task data and physiological data. Through algorithms such as deep learning, the dynamic changes of the data are analyzed to identify the changing trend of cognitive ability, timely capture cognitive fluctuations, and conduct personalized assessment of the elderly according to the results of data analysis. The assessment results are accurate and effective. According to the assessment results, a personalized training plan is formulated to achieve precise adaptation of training items and training difficulty, which is conducive to early detection of cognitive decline in the elderly and computational intervention, and effective health management.

[0054] It should be noted that as Figure 3 shown, in the elderly health management detection system, the elderly complete system tasks through touch screen interaction. The system collects behavioral task data, and at the same time, the non-invasive device equipped with the system captures the physiological data of the user, encrypts the collected data and conducts cloud transmission and integrity verification; The system preprocesses the collected data, including data cleaning, normalization, removing outliers and noise, then extracts the data features of the collected data and performs fusion, and inputs them into the cognitive state assessment model set for integrated analysis, and dynamically identifies and evaluates the cognitive state of the elderly in combination with artificial intelligence algorithms; Adjust the training frequency according to the cognitive ability score of the elderly and control the training difficulty in a stepped manner to generate a training plan, and perform real-time adjustment of the difficulty. Then, perform three-level downgrading and training reinforcement according to heart rate variability HRV and task completion; The training results are fed back to the elderly in real time. The system rewards the elderly according to their training performance and provides personalized guidance suggestions, identifies incorrect training patterns, and pushes training skills to the elderly for incorrect training patterns.

[0055] Next, refer to Figure 4 , which shows a schematic structural diagram of a computer system of an electronic device suitable for implementing the embodiments of the present application. Figure 4 The electronic device shown is only an example and should not bring any limitations to the functions and usage scope of the embodiments of the present application.

[0056] As Figure 4 shown, the computer system includes a central processing unit (CPU) 401, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 402 or the program loaded from the storage section 408 into the random access memory (RAM) 403. In the RAM 403, various programs and data required for system operation are also stored. The CPU 401, ROM 402, and RAM 403 are connected to each other through a bus 404. The input / output (I / O) interface 405 is also connected to the bus 404.

[0057] The following components are connected to the I / O interface 405: an input section 406 including a keyboard, a mouse, etc.; an output section 407 including such as a liquid crystal display (LCD) etc. and a speaker etc.; a storage section 408 including a hard disk etc.; and a communication section 409 including a network interface card such as a LAN card, a modem, etc. The communication section 409 performs communication processing via a network such as the Internet. A drive 410 is also connected to the I / O interface 405 as needed. A removable medium 411, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is mounted on the drive 410 as needed so that a computer program read out therefrom is installed into the storage section 408 as needed.

[0058] In particular, according to embodiments of the present disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product that includes a computer program carried on a computer-readable storage medium, and the computer program includes program code for performing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 409 and / or installed from the removable medium 411. When the computer program is executed by the central processing unit (CPU) 401, the above-described functions defined in the methods of the present application are performed. It should be noted that the computer-readable storage medium of the present application can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium can include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, the computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present application, the computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries the computer-readable program code. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium can also be any computer-readable storage medium other than the computer-readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable storage medium can be transmitted by any appropriate medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination of the above.

[0059] Computer program code for performing the operations of this application can be written in one or more programming languages or combinations thereof. The programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (for example, by connecting through the Internet using an Internet service provider).

[0060] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in the flowchart or block diagram can represent a module, a program segment, or a part of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks can occur in a different order than marked in the accompanying drawings. For example, two consecutive blocks shown can actually be executed substantially in parallel, and they can sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and the combinations of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0061] The modules described in the embodiments of this application can be implemented in software or in hardware.

[0062] As another aspect, the present application also provides a computer-readable storage medium, which may be included in the electronic device described in the above embodiments; or may exist alone without being assembled into the electronic device. The above computer-readable storage medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device is enabled to: collect the behavioral task data and physiological data of the elderly through a scale or non-invasive device, and transmit the behavioral task data and the physiological data to the cloud; extract data features from the behavioral task data and physiological data and perform fusion, where the data features include task completion status, reaction time, and visual behavior patterns; weight the ensemble learning model, deep learning model, and statistical learning model according to a certain weight and construct a cognitive state evaluation model set, train the cognitive state evaluation model set using a differential learning strategy, analyze the importance of the data features through a decision tree model, quantify the correlation strength between the cognitive level of the elderly and the data features through a regression model, and use the trained cognitive state evaluation model set to score the behavioral task data and physiological data to obtain a cognitive ability score; construct a dynamic evaluation system based on the cognitive ability score, determine the evaluation frequency through the cognitive ability score, determine the evaluation items using the correlation priority strategy of the data features, establish a non-linear mapping rule, enable a simplified evaluation process when the cognitive ability score is lower than a preset threshold, and add high-order executive function test items when the cognitive ability score is higher than the preset threshold to form a stepped evaluation plan; set training benchmark parameters according to the cognitive ability score, continuously track the behavioral task data and physiological data of the elderly, formulate a personalized cognitive training plan based on the dynamic evaluation system and the individual data features of the elderly, compare the training performance of the elderly and provide feedback through personalized guidance suggestions.

[0063] Finally, it should be noted that the above description is only a preferred embodiment of the present application and an explanation of the applied technical principles. Those skilled in the art should understand that the scope of the invention involved in the present application is not limited to the technical solutions formed by the specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above inventive concept. For example, the technical solutions formed by mutually replacing the above features with the (but not limited to) technical features with similar functions disclosed in the present application.

Claims

1. A health management detection method for the elderly, characterized in that, Including: S1: Collect the behavioral task data and physiological data of the elderly through a scale or non-invasive device, and transmit the behavioral task data and the physiological data to the cloud; S2: Extract data features from the behavioral task data and physiological data and fuse them. The data features include task completion, reaction time, and visual behavior patterns; S3: Weight the ensemble learning model, deep learning model, and statistical learning model according to certain weights to construct a cognitive state evaluation model set. Adopt a differential learning strategy to train the cognitive state evaluation model set. Analyze the importance of the data features through a decision tree model. Quantify the correlation strength between the cognitive level of the elderly and the data features through a regression model. Use the trained cognitive state evaluation model set to score the behavioral task data and physiological data to obtain a cognitive ability score; S4: Based on the cognitive ability score, construct a dynamic evaluation system. Determine the evaluation frequency through the cognitive ability score. Adopt the correlation degree priority strategy of the data features to determine the evaluation items. Establish a non-linear mapping rule. When the cognitive ability score is lower than a preset threshold, enable a simplified evaluation process. When the cognitive ability score is higher than the preset threshold, add high-order executive function test items to form a stepped evaluation plan; S5: Set training benchmark parameters according to the cognitive ability score. Real-time track the behavioral task data and physiological data of the elderly. Develop a personalized cognitive training plan based on the dynamic evaluation system and the individual data features of the elderly. Compare the training performance of the elderly and provide feedback through personalized guidance suggestions.

2. The method according to claim 1, characterized in that The behavioral task data includes graphic recognition data, memory task data, scale data, interface operation trajectory data, task completion time data, and attention test data; The physiological data includes blood oxygen saturation data, electroencephalogram signal data, heart rate data, and eye movement data.

3. The method according to claim 2, wherein Step S3 further includes using a support vector machine (SVM) to process the electroencephalogram signal data and blood oxygen saturation data, processing non-linearly separable problems through kernel function transformation, and analyzing the correlation between eye movement data and scale data.

4. The method according to claim 1, wherein Step S3 further includes using a naive Bayes classifier to make a conditional independence assumption for the task completion time data and interface operation trajectory data and perform rapid inference.

5. The method according to claim 1, characterized in that, Step S3 further includes calculating the Gain value for the eye movement data and scale data, sorting the importance of the data features, and screening key indicators.

6. The method according to claim 1, characterized in that Step S3 further includes using a time series algorithm to capture the long-range dependence relationship of the heart rate data through a memory gating mechanism and reduce the mean absolute error.

7. The method according to claim 1, wherein Step S5 further includes adjusting the personalized cognitive training plan through reinforcement learning and a state-action-reward mechanism.

8. An elderly health management detection system, characterized in that, Including: Acquisition module: Collect the behavioral task data and physiological data of the elderly through a scale or non-invasive device, and transmit the behavioral task data and the physiological data to the cloud; Fusion module: Extract data features from the behavioral task data and physiological data and fuse them. The data features include task completion, reaction time, and visual behavior patterns; Analysis module: Weight the ensemble learning model, deep learning model, and statistical learning model according to certain weights and construct a set of cognitive state evaluation models. Use a differential learning strategy to train the set of cognitive state evaluation models. Analyze the importance of the data features through a decision tree model, quantify the correlation strength between the cognitive level of the elderly and the data features through a regression model, and use the trained set of cognitive state evaluation models to score the behavioral task data and physiological data to obtain a cognitive ability score; Evaluation module: Based on the cognitive ability score, construct a dynamic evaluation system. Determine the evaluation frequency through the cognitive ability score, adopt the correlation degree priority strategy of the data features to determine the evaluation items, establish a non-linear mapping rule. When the cognitive ability score is lower than a preset threshold, enable a simplified evaluation process. When the cognitive ability score is higher than the preset threshold, add high-order executive function test items to form a stepped evaluation plan; Feedback module: Set training benchmark parameters according to the cognitive ability score, continuously track the behavioral task data and physiological data of the elderly, formulate a personalized cognitive training plan based on the dynamic evaluation system and the individual data features of the elderly, compare the training performance of the elderly and provide feedback through personalized guidance suggestions.

9. A computer program product having one or more computer programs thereon, characterized in that, When the computer program is executed by a computer processor, the method described in any one of claims 1-7 is implemented.

Images