Personalized rehabilitation scheme auxiliary decision-making method and system

Abstract

The invention provides a personalized rehabilitation scheme auxiliary decision-making method and system, and relates to the technical field of smart medical treatment, and the method comprises the steps: receiving the state monitoring information of a wearing user; obtaining a first functional disability condition evaluation result; obtaining a second functional disability condition evaluation result; recommended sleep posture information is obtained; obtaining a recommended rehabilitation exercise scheme; and sending the recommended sleep posture information and the recommended rehabilitation exercise scheme to an AKI patient management end. By means of the method and device, the technical problem that in the prior art, due to the fact that gradually-changing health conditions of the patient in the rehabilitation process are not fully considered, multiple health problems of the patient cannot be accurately and individually considered, and the rehabilitation scheme is possibly insufficient in pertinence can be solved; more scientific and personalized rehabilitation guidance is provided for the users in different health states, and the technical effect of improving the applicability of the rehabilitation scheme and the users is achieved.

Description

Technical Field

[0001] The present application relates to the field of smart medical technology, and in particular to a method and system for assisting decision-making on personalized rehabilitation programs. Background Art

[0002] With the advent of an aging society, the elderly population is facing challenges from a variety of chronic diseases and comorbidities. Among them, acute kidney injury (AKI), as a common clinical symptom, has become one of the leading causes of death and disability worldwide. AKI not only has a serious impact on the patient's kidney function, but is also often accompanied by the decline of other organ functions, leading to multi-system dysfunction. Therefore, the rehabilitation and care of AKI patients, especially elderly patients with comorbid AKI, requires comprehensive consideration of the functional status of each organ and the development of personalized and precise rehabilitation plans.

[0003] With the development of medical technology, intelligent rehabilitation decision-making support systems have emerged. By comprehensively considering the patient's vital signs, disease conditions, and physiological function data, they assist medical personnel in formulating personalized rehabilitation plans, thereby improving the rehabilitation effect and the patient's quality of life. Traditional rehabilitation plans are usually static and fail to fully consider the patient's gradually changing health status during the rehabilitation process. For elderly patients with comorbid AKI, due to the presence of multiple complications, such as heart disease and respiratory problems, traditional plans cannot accurately and individually consider the patient's multiple health problems, which may lead to insufficient targeted rehabilitation plans. Summary of the invention

[0004] The purpose of this application is to provide a personalized rehabilitation program decision-making support method and system to solve the technical problems in the prior art that the rehabilitation program is not targeted enough due to the failure to fully consider the patient's gradually changing health status during the rehabilitation process and the inability to accurately and individually consider the patient's multiple health problems.

[0005] In view of the above problems, the present application provides a method and system for assisting decision-making of personalized rehabilitation programs.

[0006] In a first aspect, the present application provides a personalized rehabilitation program decision-making assistance method, which is implemented by a personalized rehabilitation program decision-making assistance system, wherein the personalized rehabilitation program decision-making assistance method includes: receiving wearable user status monitoring information from a wearable monitoring device, wherein the wearable user status monitoring information includes daily activity status monitoring information and rehabilitation exercise status monitoring information; performing functional disability status analysis based on sleep status monitoring information, respiratory status monitoring information and swallowing status monitoring information of the daily activity status monitoring information to obtain a first functional disability status assessment result; collecting heart, lung, brain and kidney status monitoring information of the rehabilitation exercise status monitoring information to perform functional disability status analysis to obtain a second functional disability status assessment result; optimizing sleeping posture based on the first functional disability status assessment result to obtain recommended sleeping posture information; optimizing exercise plan based on the second functional disability status assessment result to obtain a recommended rehabilitation exercise plan; sending the recommended sleeping posture information and the recommended rehabilitation exercise plan to the management end of elderly patients with comorbid AKI.

[0007] In a second aspect, the present application also provides a personalized rehabilitation program decision-making support system for executing a personalized rehabilitation program decision-making support method as described in the first aspect, wherein the personalized rehabilitation program decision-making support system includes: a user state monitoring module, which is used to receive wearable user state monitoring information from a wearable monitoring device, wherein the wearable user state monitoring information includes daily activity state monitoring information and rehabilitation exercise state monitoring information; a first functional disability analysis module, which is used to perform functional disability status analysis based on sleep state monitoring information, respiratory state monitoring information and swallowing state monitoring information of the daily activity state monitoring information to obtain a first functional disability status assessment result; a second functional disability analysis module, which is used to collect heart, lung, brain and kidney state monitoring information of the rehabilitation exercise state monitoring information to perform functional disability status analysis to obtain a second functional disability status assessment result; a sleep posture optimization module, which is used to optimize the sleep posture according to the first functional disability status assessment result to obtain recommended sleep posture information; an exercise plan optimization module, which is used to optimize the exercise plan according to the second functional disability status assessment result to obtain a recommended rehabilitation exercise plan; a plan sending module, which is used to send the recommended sleep posture information and the recommended rehabilitation exercise plan to the management end of the elderly comorbid AKI patient.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages: Receive wearable user status monitoring information from a wearable monitoring device, wherein the wearable user status monitoring information includes daily activity status monitoring information and rehabilitation exercise status monitoring information; perform functional disability status analysis based on the sleep status monitoring information, respiratory status monitoring information, and swallowing status monitoring information of the daily activity status monitoring information to obtain a first functional disability status assessment result; collect the heart, lung, brain, and kidney status monitoring information of the rehabilitation exercise status monitoring information to perform functional disability status analysis to obtain a second functional disability status assessment result; optimize the sleeping posture based on the first functional disability status assessment result to obtain recommended sleeping posture information; optimize the exercise plan based on the second functional disability status assessment result to obtain a recommended rehabilitation exercise plan; send the recommended sleeping posture information and the recommended rehabilitation exercise plan to the management end of elderly comorbid AKI patients. Through real-time collection and analysis of multi-source data of users, functional disability status assessment is performed, sleep and exercise plans are adjusted in a personalized manner, and more scientific and personalized rehabilitation guidance is provided for users in different health states, thereby achieving the technical effect of improving the applicability of rehabilitation plans and users.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0011] Figure 1 A flowchart of a personalized rehabilitation program decision-making support method for this application.

[0012] Figure 2 This is a structural diagram of a personalized rehabilitation program decision support system for this application.

[0013] Explanation of the reference numerals: user status monitoring module 11 , first functional disability analysis module 12 , second functional disability analysis module 13 , sleeping posture optimization module 14 , exercise plan optimization module 15 , plan sending module 16 . DETAILED DESCRIPTION

[0014] This application provides a personalized rehabilitation program decision-making assistance method and system to solve the technical problem in the prior art that the rehabilitation program is not targeted enough due to the failure to fully consider the patient's gradually changing health status during the rehabilitation process and the inability to accurately and individually consider the patient's multiple health problems. Through real-time collection and analysis of multi-source data of users, functional disability status assessment, personalized adjustment of sleep and exercise plans, and providing more scientific and personalized rehabilitation guidance for users in different health conditions, the technical effect of improving the applicability of rehabilitation programs and users is achieved.

[0015] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0016] For example, please refer to the attached Figure 1 The present application provides a personalized rehabilitation program decision-making assistance method, wherein the personalized rehabilitation program decision-making assistance method is applied to a personalized rehabilitation program decision-making assistance system, and the personalized rehabilitation program decision-making assistance method specifically includes the following steps: Step 1: receiving wearer status monitoring information from a wearable monitoring device, wherein the wearer status monitoring information includes daily activity status monitoring information and rehabilitation exercise status monitoring information.

[0017] Specifically, through the communication connection with the wearable monitoring device, it is possible to receive the wearable user status monitoring information, wherein the wearable monitoring device usually includes various sensors, such as vibration sensors, accelerometers, gyroscopes, heart rate sensors, etc., which can collect and transmit the user's physiological data in real time. In this application, the vibration state of the user is collected by the vibration sensor of the wearable monitoring device, and then the vibration characteristics are analyzed to determine the user's daily activity status monitoring information or rehabilitation exercise status monitoring information, and the wearable user status monitoring information is generated by the daily activity status monitoring information or rehabilitation exercise status monitoring information.

[0018] In layman's terms, the monitoring parameters corresponding to daily activity status monitoring information and rehabilitation exercise status monitoring information are different. The monitoring parameters corresponding to daily activity status monitoring information are the user's activity data in daily life, including sleep status, breathing status, swallowing status, etc., which can reflect the user's activity level and whether there are life obstacles. Rehabilitation exercise status monitoring information is the user's exercise performance during the rehabilitation process, including heart, lung, brain and kidney status parameters during rehabilitation training. By integrating daily activity status and rehabilitation exercise status monitoring information, we can fully understand the user's activity ability and rehabilitation progress, and provide data support for the subsequent optimization of sleep and rehabilitation plans.

[0019] Step 2: Perform functional disability status analysis based on the sleep status monitoring information, the respiratory status monitoring information, and the swallowing status monitoring information of the daily activity status monitoring information to obtain a first functional disability status assessment result.

[0020] Specifically, sleep state monitoring information is usually obtained through wearable monitoring devices such as smart bracelets and smart watches, and sleep state monitoring information is obtained by monitoring the user's sleep quality, sleep cycle, and the ratio of deep sleep to light sleep. Respiratory state monitoring information is usually collected through smart chest straps or respiratory monitoring devices in wearable monitoring devices, which record the user's respiratory rate, respiratory depth, oxygen saturation and other data to form respiratory state monitoring information. Swallowing state monitoring information is a key data for assessing whether the patient has swallowing difficulties (such as swallowing disorders or swallowing function decline). It is usually collected through special swallowing monitoring devices in wearable monitoring devices. Swallowing state monitoring information may include swallowing frequency, swallowing duration, muscle activity during swallowing (electromyography data), etc.

[0021] By integrating the sleep state monitoring information, respiratory state monitoring information and swallowing state monitoring information, the user's functional disability status is analyzed by establishing a respiratory disability status assessment model, and the user's functional disability type is inferred as the first functional disability status assessment result. For example, if the analysis shows that the first functional disability status assessment result of an elderly patient is moderate respiratory dysfunction, accompanied by mild swallowing disorder, this means that the user's respiratory function is impaired to a certain extent. Through this multi-dimensional data analysis, not only can the patient's functional status be accurately assessed, but also a detailed basis can be provided for personalized sleep plans.

[0022] Step 3: Collect the heart, lung, brain and kidney status monitoring information of the rehabilitation exercise status monitoring information to perform functional disability status analysis and obtain the second functional disability status assessment result.

[0023] Specifically, the heart, lung, brain and kidney status monitoring information of the rehabilitation exercise status monitoring information includes monitoring data corresponding to the heart, lung, brain and kidney, including heart rate, electrocardiogram, blood pressure and other heart health related data, lung capacity, blood oxygen saturation, respiratory rate and other lung health related parameters, EEG, nerve reflex test data and other brain health related parameters, as well as urine information, kidney ultrasound data, creatinine and urea nitrogen data in the blood and other kidney health related parameters. The same method as that used to obtain the first functional disability status assessment results is used to train the organ disability status assessment model, analyze the heart, lung, brain and kidney status monitoring information, integrate the health status data of each organ, and assess the organ disability type as the second functional disability status assessment result, thereby providing a basis for subsequent rehabilitation treatment and helping to achieve a more accurate and personalized rehabilitation plan.

[0024] Step 4: Optimize sleeping posture according to the first functional disability status assessment result to obtain recommended sleeping posture information.

[0025] Specifically, the first functional disability status assessment result is the user's respiratory disability type. The core goal of sleep posture optimization is to improve the patient's rest quality and reduce possible functional disabilities through scientific sleep posture adjustment. For example, for patients with respiratory problems (such as apnea, shortness of breath, etc.), side-lying or semi-lying postures are recommended to reduce airway obstruction and promote smooth breathing. For patients with poor swallowing function, it is recommended to raise the head appropriately to avoid the risk of aspiration or suffocation. Specifically, by analyzing the user's sleep posture records in the large data set, the effective sleeping postures adopted by users with similar health conditions can be found as recommended sleeping posture information, which can help users get higher quality sleep and thus prevent the deterioration of respiratory disabilities.

[0026] Step 5: Optimize the exercise plan based on the second functional disability status assessment results to obtain a recommended rehabilitation exercise plan.

[0027] Specifically, the result of the second functional disability status assessment is the user's organ disability type. The core goal of the exercise plan is to develop a more targeted rehabilitation exercise plan based on the health status of the patient's heart, lungs, brain, kidneys and other important organs, avoid the health risks caused by excessive exercise or improper exercise, and maximize the rehabilitation effect. According to the result of the second functional disability status assessment, the type, intensity and duration of exercise are reasonably selected to help patients improve or maintain cardiopulmonary function, promote blood circulation, enhance muscle strength, and improve overall health. When optimizing the exercise plan, the rehabilitation exercise plan record data set of the same cluster users can be collected according to the second functional disability assessment result and the user's medical record information, that is, the rehabilitation exercise plan set of other users with the same medical record and the same organ function disability type is obtained. Further statistics are made on the trigger frequency of any rehabilitation exercise plan in the rehabilitation exercise plan record data set of the same cluster users, that is, the proportion of any rehabilitation exercise plan in the rehabilitation exercise plan record data set of the same cluster users, and the rehabilitation exercise plan with a trigger frequency greater than or equal to the trigger frequency threshold is selected as the recommended rehabilitation exercise plan, providing users with scientific and personalized exercise guidance, which can not only help patients restore their physical functions, but also reduce the risks in exercise and promote their comprehensive rehabilitation.

[0028] Step 6: Send the recommended sleeping posture information and the recommended rehabilitation exercise program to the management terminal of elderly patients with comorbid AKI.

[0029] Specifically, the elderly comorbid AKI patient management terminal serves as an interactive interface between the user and the rehabilitation system, which can receive, store and display rehabilitation suggestions, and provide users with functional support for execution and feedback. Sending the recommended sleeping posture information and the recommended rehabilitation exercise plan to the elderly comorbid AKI patient management terminal requires the use of modern medical information systems and communication technologies. Before sending, the recommended information is formatted, such as generating a structured data file (such as JSON or XML format), and using encryption technology (such as AES encryption) to protect data privacy. The data is sent to the elderly comorbid AKI patient management terminal through existing secure communication protocols (such as HTTPS or MQTT protocols). The transmission process will ensure the integrity and security of the data. The elderly comorbid AKI patient management terminal intuitively displays the recommended sleeping posture information and recommended rehabilitation exercise plans in the form of pictures and texts. For example, sleep suggestions may be presented in the form of human posture diagrams, and exercise plans are displayed in the form of daily plans or step-by-step instructions. Patients can obtain scientific and personalized rehabilitation guidance in a timely manner, while greatly improving the operability of the plan.

[0030] Furthermore, step one of this application includes: The vibration sensor of the wearable monitoring device is used to collect user vibration status monitoring information; time domain features are extracted from the user vibration status monitoring information to obtain user vibration time domain features; Fourier transform is performed on the user vibration status monitoring information to obtain user vibration frequency domain features; activity types are classified according to the user vibration frequency domain features and the user vibration time domain features to obtain activity type calibration results; wherein the activity type calibration result is one of the daily activity status monitoring information and the rehabilitation exercise status monitoring information; and classified collection of the wearable user status monitoring information is performed according to the activity type calibration results.

[0031] Specifically, the vibration sensor provides valuable physiological data by sensing the tiny vibration changes of the user during activities or movements, which can reflect the user's exercise intensity, frequency and amplitude in different states. Vibration state monitoring information, including vibration amplitude, frequency and other data, can be directly collected through the vibration sensor. Time domain feature extraction refers to extracting basic information such as amplitude, frequency, waveform, etc. from the vibration signal as the user's vibration time domain feature. For example, if the vibration frequency of a user during activity shows a certain periodic pattern, its frequency characteristics will be recorded, which may be expressed as the number of vibrations per second. Then, the user's vibration state monitoring information is Fourier transformed to obtain the user's vibration frequency domain characteristics. Fourier transform is an existing mathematical tool that can convert time domain signals into frequency domain signals and reveal the frequency components of the signal. In the frequency domain of the vibration signal, peaks of different frequencies can often be seen, which are used as the user's vibration frequency domain characteristics. For example, the low-frequency area may correspond to a stationary state or slow activity, while the high-frequency area represents fast movement or high-intensity activity. This frequency domain feature can effectively reflect the intensity and pattern of the movement and provide a basis for the subsequent classification of activity types.

[0032] By combining the user's vibration frequency domain features and the user's vibration time domain features, the user's activity type can be classified. The process of activity type classification usually involves machine learning algorithms, such as using support vector machines (SVM), decision trees or neural networks to compare and train the extracted features with pre-labeled activity type data. Through training, it can be determined whether the user's current activity type belongs to daily activity status (such as walking, standing, sitting, sleeping, etc.) or rehabilitation exercise status (such as gait training, balance exercises, strength training, etc.). Finally, according to the activity type calibration results, the user's wearing status monitoring information is classified and collected to ensure that different types of exercise status can be accurately monitored, and corresponding interventions or suggestions are made as needed.

[0033] Furthermore, the present application also includes the following steps: Collect user vibration state recording data and activity type identification data; perform time domain feature extraction on the user vibration state recording data to obtain user vibration state time domain recording features; perform Fourier transform on the user vibration state recording data to obtain user vibration state frequency domain recording features; use the user vibration state time domain recording features and the user vibration state frequency domain recording features as input and the activity type identification data as supervision to train an activity type classifier.

[0034] Specifically, the user vibration state recording data and activity type identification data are sample data obtained through testing by professional and technical personnel in this field. The vibration state data corresponding to different activity types of different users are obtained through testing and data annotation is obtained. The vibration state recording data records the vibration intensity and frequency changes at each moment. Based on the corresponding user's activity type, data annotation is performed to obtain the corresponding activity type identification data. Further, the user vibration state recording data is subjected to time domain feature extraction. The time domain features include but are not limited to statistics such as the mean, variance, maximum value, minimum value, and peak value of the signal, which can reveal the basic trend and fluctuation of the vibration signal, and obtain the user's vibration state time domain recording characteristics. Next, the user vibration state recording data is subjected to Fourier transform and converted into frequency domain features, revealing the different frequency components and their intensities in the signal, and obtaining the user's vibration state frequency domain recording characteristics.

[0035] The user's vibration state time domain recording features and the user's vibration state frequency domain recording features are used as inputs, and supervised learning is performed using activity type identification data to train an activity type classifier. The activity type classifier can be based on existing machine learning models, such as neural networks, support vector machines and other components. The goal is to accurately predict the user's activity type in new uncalibrated data by learning the relationship between the time domain and frequency domain features extracted from the vibration data and the calibrated activity types. For example, 5,000 user vibration recording data were used in the experiment, and these data were paired with the calibrated activity types. After multiple iterations of training, a classification model that can accurately distinguish between activities such as walking, sitting, and running was finally constructed. The accuracy during the training process gradually increased with the optimization of the training, and eventually reached an activity classification accuracy of 95%.

[0036] Finally, the obtained user vibration frequency domain features and user vibration time domain features are input into the activity type classifier for analysis, and the activity type calibration results are output to provide more accurate data support for the user's personalized rehabilitation plan.

[0037] Further, step 2 of this application includes: The sleep state monitoring information, the respiratory state monitoring information and the swallowing state monitoring information are input into a respiratory disability status assessment model for analysis to obtain a respiratory disability type, which is set as the first functional disability status assessment result.

[0038] Specifically, comprehensive analysis of sleep state monitoring information, respiratory state monitoring information, and swallowing state monitoring information is crucial to assessing the patient's respiratory disability status. These data can reflect the patient's respiratory function and swallowing ability during daily activities and rehabilitation. After integrating these monitoring information, they are input into the respiratory disability status assessment model for further analysis. The respiratory disability status assessment model is usually trained through a variety of algorithms, combining the user's sleep, breathing, and swallowing information to determine the type of respiratory disability, which is set as the first functional disability status assessment result. For example, machine learning technology may be used to learn the clinical characteristics of different patients through a large amount of training data to determine the type of respiratory disability. Based on the evaluation results, personalized rehabilitation plans can be further provided, such as improving sleeping posture to improve the user's respiratory function and reduce the impact of respiratory disability on daily life.

[0039] Furthermore, the present application also includes the following steps: Construct a multi-classification loss function and an integrated loss function; train several respiratory disability condition assessment sub-classifiers according to the multi-classification loss function; train the fully connected weights of the several respiratory disability condition assessment sub-classifiers according to the integrated loss function; fully connect the several respiratory disability condition assessment sub-classifiers according to the fully connected weights to obtain the respiratory disability condition assessment model.

[0040] Specifically, the multi-classification loss function includes multiple existing loss functions, such as the cross entropy loss function, the Hinge loss function (used for support vector machines), the adversarial loss function, etc. The technicians in this field can select multiple existing loss functions by themselves. They are prior art and will not be expanded here one by one. The multiple loss functions can be the loss functions used in the training of different machine learning models. Further, several respiratory disability status assessment sub-classifiers can be trained according to the multi-classification loss function, that is, one loss function is used to train a respiratory disability status assessment sub-classifier, and the functions of several respiratory disability status assessment sub-classifiers are the same, all of which are based on the user's sleep, breathing and swallowing information to analyze the corresponding respiratory disability type, but there may be differences in accuracy.

[0041] Under the framework of the multi-classification loss function, several respiratory disability condition assessment sub-classifiers are trained. Specifically, professional and technical personnel in this field obtain a large amount of corresponding sleep state monitoring information sample data, respiratory state monitoring information samples, swallowing state monitoring information samples and respiratory disability type annotation data based on the existing technology. These sample data can be grouped to obtain multiple groups of training samples. Based on the multi-classification loss function and the existing machine learning model, multiple groups of training samples are used for training and testing respectively to obtain several respiratory disability condition assessment sub-classifiers that are trained to convergence.

[0042] Next, several respiratory disability status assessment sub-classifiers are connected through an integrated loss function to improve the stability and accuracy of the prediction. The integrated loss function usually weights the losses of several respiratory disability status assessment sub-classifiers, that is, multiple groups of connection weights can be randomly generated, each group of connection weights includes the weights corresponding to several respiratory disability status assessment sub-classifiers, and the sum of the weights of several respiratory disability status assessment sub-classifiers is 1, that is, based on the integrated loss function, the losses of several respiratory disability status assessment sub-classifiers are comprehensively calculated according to their respective weights, and a set of connection weights with the smallest integrated loss is obtained as the final full connection weight of several respiratory disability status assessment sub-classifiers, thereby ensuring the performance of the overall model. The full connection weight refers to the weighted relationship between several respiratory disability status assessment sub-classifiers at the time of output, which determines how the final prediction result is integrated from the output of each sub-classifier.

[0043] Full connection is a common layer structure in neural networks. The outputs of several respiratory disability status assessment sub-classifiers are connected together through weights to form a unified prediction model, that is, the respiratory disability status assessment model, which can comprehensively judge the patient's respiratory disability type based on the input multi-source data (such as sleep status, breathing status, swallowing status, etc.), and provide model support for the assessment of functional disability status.

[0044] Furthermore, the present application also includes the following steps: The integrated loss function is: ; ; in, Represents the first integrated loss on a set of connection weights, The second integrated loss that represents a set of connection weights, represents the loss of the i-th classifier, Characterizes the sub-classifier loss threshold, Characterize the weighted mean weight of the i-th classifier; based on the first integrated loss threshold and the second integrated loss threshold, combined with the integrated loss function, train the fully connected weights of the several respiratory disability condition assessment sub-classifiers.

[0045] Specifically, in the above integrated loss function, Represents the loss of the ith classifier, that is, the final loss value obtained after the ith classifier is trained to convergence through the multi-classification loss function and the corresponding training samples. The loss threshold that characterizes the sub-classifier, that is, a preset loss threshold, can be understood as an expected loss threshold under ideal conditions, which is specifically set by professional and technical personnel in this field. The weighted mean weight that represents the i-th classifier, that is, the weight of the i-th classifier in multiple sets of randomly generated connection weights, is also a target that needs to be trained and optimized through the integrated loss function. is the symbol for standard deviation, , N is the total number of several respiratory disability status assessment sub-classifiers.

[0046] The first integrated loss threshold and the second integrated loss threshold are the expected first integrated loss and the expected second integrated loss set by professionals in the field, such as 0.01. Multiple groups of connection weights are randomly generated, and the multiple groups of connection weights are calculated by the integrated loss function to obtain a group of connection weights that meet the first integrated loss threshold and the second integrated loss threshold as the full connection weights of several respiratory disability status assessment sub-classifiers.

[0047] Specifically, the integrated loss function can also be used as a fitness function, and satisfying the first integrated loss threshold and the second integrated loss threshold can be used as the iteration stop condition. The link weights can be iteratively optimized through existing optimization algorithms, such as particle swarm optimization algorithm, genetic algorithm, etc., so as to obtain the full connection weight that satisfies the first integrated loss threshold and the second integrated loss threshold. The optimization algorithm is a common technical means used by technicians in this field and will not be expanded here.

[0048] Similarly, the same method as that used to obtain the first functional disability status assessment result can be used to construct an organ disability status assessment model, analyze the heart, lung, brain, and kidney status monitoring information, and evaluate the organ disability type as the second functional disability status assessment result. Specifically, a multi-classification loss function and an integrated loss function are also used to obtain sample data of the heart, lung, brain, and kidney status monitoring and the corresponding organ disability type as sample data, and the sample data is divided. According to the division results, a multi-classification loss function is used to train several sub-classifiers, and then the integrated loss function is used to train the fully connected weights of several sub-classifiers, and several sub-classifiers are fully connected to obtain an organ disability status assessment model, and then the heart, lung, brain, and kidney status monitoring information is analyzed to evaluate the organ disability type, that is, the second functional disability status assessment result.

[0049] Further, step 4 of this application includes: Based on the first functional disability status assessment result and combined with the user's medical history information, a data set of sleep posture records of users in the same cluster is collected; a trigger frequency set of the sleep posture record data set of users in the same cluster is extracted; and a sleep posture whose trigger frequency set is greater than or equal to a trigger frequency threshold is selected and added to the recommended sleep posture information.

[0050] Specifically, the user's medical history information is first collected, such as the user's past medical history, chronic diseases, surgical records and other information, to form the user's medical history information. According to the first functional disability assessment result and the user's medical history information, based on the existing big data technology, the sleeping postures of other users with the same first functional disability assessment result and the same user's medical history information are collected to form a cluster user sleeping posture record data set. The users in the cluster user sleeping posture record data set have similar respiratory disability type problems and similar medical history information, which makes the analysis more accurate. The next step is to conduct an in-depth analysis of the cluster user sleeping posture record data set, extract the trigger frequency of various types of sleeping postures, and form a trigger frequency set. The trigger frequency refers to the proportion of a certain specific sleeping posture in the cluster user group. For example, the cluster user sleeping posture record data set has 1,000 users' sleeping postures, among which the side-lying posture appears 800 times, then the corresponding trigger frequency is 0.8, which reflects the effectiveness of the posture in patients with a certain type of specific health problems. The trigger frequency threshold is set by technicians in this field, such as 0.7, and the sleeping posture with a trigger frequency greater than or equal to the trigger frequency threshold is selected and added to the recommended sleeping posture information. Through this data-driven approach, sleep posture recommendations are not only supported by data, but can also be precisely optimized based on the patient's actual needs to maximize the sleep adjustment effect.

[0051] Similarly, the same method can be used to collect a data set of rehabilitation exercise plan records for users in the same cluster based on the second functional disability assessment results and user medical history information, further count the triggering frequency of any rehabilitation exercise plan in the data set of rehabilitation exercise plan records for users in the same cluster, and select the rehabilitation exercise plan with a trigger frequency greater than or equal to the trigger frequency threshold as the recommended rehabilitation exercise plan to provide users with scientific and personalized exercise guidance.

[0052] In summary, the personalized rehabilitation program decision-making assistance method provided by this application has the following technical effects: Receive wearable user status monitoring information from a wearable monitoring device, wherein the wearable user status monitoring information includes daily activity status monitoring information and rehabilitation exercise status monitoring information; perform functional disability status analysis based on the sleep status monitoring information, respiratory status monitoring information, and swallowing status monitoring information of the daily activity status monitoring information to obtain a first functional disability status assessment result; collect the heart, lung, brain, and kidney status monitoring information of the rehabilitation exercise status monitoring information to perform functional disability status analysis to obtain a second functional disability status assessment result; optimize the sleeping posture based on the first functional disability status assessment result to obtain recommended sleeping posture information; optimize the exercise plan based on the second functional disability status assessment result to obtain a recommended rehabilitation exercise plan; send the recommended sleeping posture information and the recommended rehabilitation exercise plan to the management end of elderly comorbid AKI patients. Through real-time collection and analysis of multi-source data of users, functional disability status assessment is performed, sleep and exercise plans are adjusted in a personalized manner, and more scientific and personalized rehabilitation guidance is provided for users in different health states, thereby achieving the technical effect of improving the applicability of rehabilitation plans and users.

[0053] Embodiment 2: Based on the same inventive concept as the personalized rehabilitation program decision-making assistance method in the above embodiment, the present application also provides a personalized rehabilitation program decision-making assistance system, please refer to the attached Figure 2 , the personalized rehabilitation program decision-making support system comprises: The user status monitoring module 11 is used to receive the wearing user status monitoring information from the wearable monitoring device, wherein the wearing user status monitoring information includes daily activity status monitoring information and rehabilitation exercise status monitoring information.

[0054] The first functional disability analysis module 12 is used to perform functional disability status analysis based on the sleep status monitoring information, the respiratory status monitoring information and the swallowing status monitoring information of the daily activity status monitoring information to obtain a first functional disability status assessment result.

[0055] The second functional disability analysis module 13 is used to collect the heart, lung, brain and kidney status monitoring information of the rehabilitation exercise status monitoring information to perform functional disability status analysis and obtain a second functional disability status assessment result.

[0056] The sleeping posture optimization module 14 is used to optimize the sleeping posture according to the first functional disability status assessment result and obtain recommended sleeping posture information.

[0057] The exercise program optimization module 15 is used to optimize the exercise program according to the second functional disability status assessment result to obtain a recommended rehabilitation exercise program.

[0058] The plan sending module 16 is used to send the recommended sleeping posture information and the recommended rehabilitation exercise plan to the management terminal of the elderly comorbid AKI patients.

[0059] Furthermore, the user status monitoring module 11 in the personalized rehabilitation program decision support system is also used for: The vibration sensor of the wearable monitoring device is used to collect user vibration status monitoring information; time domain features are extracted from the user vibration status monitoring information to obtain user vibration time domain features; Fourier transform is performed on the user vibration status monitoring information to obtain user vibration frequency domain features; activity types are classified according to the user vibration frequency domain features and the user vibration time domain features to obtain activity type calibration results; wherein the activity type calibration result is one of the daily activity status monitoring information and the rehabilitation exercise status monitoring information; and classified collection of the wearable user status monitoring information is performed according to the activity type calibration results.

[0060] Furthermore, the user status monitoring module 11 in the personalized rehabilitation program decision support system is also used for: Collect user vibration state recording data and activity type identification data; perform time domain feature extraction on the user vibration state recording data to obtain user vibration state time domain recording features; perform Fourier transform on the user vibration state recording data to obtain user vibration state frequency domain recording features; use the user vibration state time domain recording features and the user vibration state frequency domain recording features as input and the activity type identification data as supervision to train an activity type classifier.

[0061] Furthermore, the first functional disability analysis module 12 in the personalized rehabilitation program decision support system is also used for: The sleep state monitoring information, the respiratory state monitoring information and the swallowing state monitoring information are input into a respiratory disability status assessment model for analysis to obtain a respiratory disability type, which is set as the first functional disability status assessment result.

[0062] Furthermore, the first functional disability analysis module 12 in the personalized rehabilitation program decision support system is also used for: Construct a multi-classification loss function and an integrated loss function; train several respiratory disability condition assessment sub-classifiers according to the multi-classification loss function; train the fully connected weights of the several respiratory disability condition assessment sub-classifiers according to the integrated loss function; fully connect the several respiratory disability condition assessment sub-classifiers according to the fully connected weights to obtain the respiratory disability condition assessment model.

[0063] Furthermore, the first functional disability analysis module 12 in the personalized rehabilitation program decision support system is also used for: The integrated loss function is: ; ; in, Represents the first integrated loss on a set of connection weights, The second integrated loss that represents a set of connection weights, represents the loss of the i-th classifier, Characterizes the sub-classifier loss threshold, Characterize the weighted mean weight of the i-th classifier; based on the first integrated loss threshold and the second integrated loss threshold, combined with the integrated loss function, train the fully connected weights of the several respiratory disability condition assessment sub-classifiers.

[0064] Furthermore, the sleeping posture optimization module 14 in the personalized rehabilitation program auxiliary decision system is also used for: Based on the first functional disability status assessment result and combined with the user's medical history information, a data set of sleep posture records of users in the same cluster is collected; a trigger frequency set of the sleep posture record data set of users in the same cluster is extracted; and a sleep posture whose trigger frequency set is greater than or equal to a trigger frequency threshold is selected and added to the recommended sleep posture information.

[0065] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Figure 1 The personalized rehabilitation program decision-making assistance method and specific examples in the first embodiment are also applicable to a personalized rehabilitation program decision-making assistance system in this embodiment. Through the above detailed description of a personalized rehabilitation program decision-making assistance method, those skilled in the art can clearly know a personalized rehabilitation program decision-making assistance system in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0066] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0067] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A personalized rehabilitation program decision-making assistance method, characterized in that: Applied to a personalized rehabilitation program decision-making support system, wherein the personalized rehabilitation program decision-making support system is communicatively connected to a wearable monitoring device, and includes: Receiving wearer status monitoring information from a wearable monitoring device, wherein the wearer status monitoring information includes daily activity status monitoring information and rehabilitation exercise status monitoring information; Perform functional disability status analysis based on the sleep status monitoring information, the breathing status monitoring information, and the swallowing status monitoring information of the daily activity status monitoring information to obtain a first functional disability status assessment result; Collect the heart, lung, brain and kidney status monitoring information of the rehabilitation exercise status monitoring information to analyze the functional disability status and obtain the second functional disability status assessment result; Optimize sleeping posture according to the first functional disability status assessment result to obtain recommended sleeping posture information; Optimizing the exercise program according to the second functional disability status assessment result to obtain a recommended rehabilitation exercise program; The recommended sleeping posture information and the recommended rehabilitation exercise program are sent to the management terminal of elderly patients with comorbid AKI.

2. A personalized rehabilitation program decision-making assistance method as claimed in claim 1, characterized in that: Receive wearable user status monitoring information from the wearable monitoring device, including: Collecting user vibration status monitoring information through the vibration sensor of the wearable monitoring device; Extracting time domain features from the user vibration state monitoring information to obtain user vibration time domain features; Performing Fourier transform on the user vibration state monitoring information to obtain user vibration frequency domain characteristics; Classify the activity type according to the user vibration frequency domain characteristics and the user vibration time domain characteristics, and obtain an activity type calibration result; Wherein, the activity type calibration result is one of the daily activity state monitoring information and the rehabilitation exercise state monitoring information; The wearer status monitoring information is collected by classification according to the activity type calibration result.

3. A personalized rehabilitation program decision-making assistance method as claimed in claim 2, characterized in that: Classifying the activity type according to the user vibration frequency domain characteristics and the user vibration time domain characteristics to obtain the activity type calibration result includes: Collect user vibration status record data and activity type identification data; Extracting time domain features from the user vibration state record data to obtain time domain record features of the user vibration state; Performing Fourier transform on the user vibration state recording data to obtain frequency domain recording features of the user vibration state; An activity type classifier is trained by taking the time domain recording features of the user's vibration state and the frequency domain recording features of the user's vibration state as input and taking the activity type identification data as supervision.

4. A personalized rehabilitation program decision-making assistance method as claimed in claim 1, characterized in that: Performing functional disability status analysis based on the sleep status monitoring information, the breathing status monitoring information, and the swallowing status monitoring information of the daily activity status monitoring information to obtain a first functional disability status assessment result includes: The sleep state monitoring information, the respiratory state monitoring information and the swallowing state monitoring information are input into a respiratory disability status assessment model for analysis to obtain a respiratory disability type, which is set as the first functional disability status assessment result.

5. A personalized rehabilitation program decision-making assistance method as claimed in claim 4, characterized in that: The sleep state monitoring information, the breathing state monitoring information and the swallowing state monitoring information are input into a respiratory disability status assessment model for analysis to obtain a respiratory disability type, which includes: Construct multi-classification loss function and integrated loss function; training a plurality of respiratory disability status assessment sub-classifiers according to the multi-classification loss function; training the fully connected weights of the plurality of respiratory disability condition assessment sub-classifiers according to the integrated loss function; The plurality of respiratory disability condition assessment sub-classifiers are fully connected according to the fully connected weights to obtain the respiratory disability condition assessment model.

6. A personalized rehabilitation program decision-making assistance method as claimed in claim 5, characterized in that: The integrated loss function is: ； ； in, Represents the first integrated loss on a set of connection weights, The second integrated loss that represents a set of connection weights, represents the loss of the i-th classifier, Characterizes the sub-classifier loss threshold, Represents the weighted mean weight of the i-th classifier; Based on the first integrated loss threshold and the second integrated loss threshold, combined with the integrated loss function, the fully connected weights of the several respiratory disability condition assessment sub-classifiers are trained.

7. The method for assisting decision-making of a personalized rehabilitation program according to claim 1, characterized in that: Optimizing the sleeping posture according to the first functional disability status assessment result to obtain recommended sleeping posture information, including: Based on the first functional disability status assessment result and in combination with the user's medical record information, a data set of sleeping posture records of users in the same cluster is collected; Extracting a trigger frequency set of the sleep posture record data set of users in the same cluster; The sleeping postures whose trigger frequency set is greater than or equal to the trigger frequency threshold are selected and added into the recommended sleeping posture information.

8. A personalized rehabilitation program decision-making support system, characterized in that: The steps for implementing the method for assisting decision-making of a personalized rehabilitation program as described in any one of claims 1 to 7, wherein the personalized rehabilitation program assisting decision-making system comprises: A user status monitoring module, configured to receive wearer status monitoring information from a wearable monitoring device, wherein the wearer status monitoring information includes daily activity status monitoring information and rehabilitation exercise status monitoring information; A first functional disability analysis module is used to perform functional disability status analysis based on the sleep status monitoring information, the breathing status monitoring information and the swallowing status monitoring information of the daily activity status monitoring information to obtain a first functional disability status assessment result; The second functional disability analysis module is used to collect the heart, lung, brain and kidney status monitoring information of the rehabilitation exercise status monitoring information to perform functional disability status analysis and obtain the second functional disability status assessment result; A sleeping posture optimization module, used to optimize the sleeping posture according to the first functional disability status assessment result and obtain recommended sleeping posture information; an exercise program optimization module, used to optimize the exercise program according to the second functional disability status assessment result to obtain a recommended rehabilitation exercise program; The program sending module is used to send the recommended sleeping posture information and the recommended rehabilitation exercise program to the management terminal of elderly patients with comorbid AKI.

Images