A rehabilitation motivation and health management system based on cloud data visualization

Abstract

The application relates to the technical field of rehabilitation medical devices, in particular to a rehabilitation encouragement and health management system based on cloud data visualization, which executes a personalized rehabilitation evaluation and encouragement process based on a TOPSIS model in the following manner during system operation: joint motor torque feedback, hub motor acceleration, gait frequency and joint motor torque feedback of bioelectric signal feedback, abnormal posture frequency, gait frequency and bioelectric signal intensity data of a user in a plurality of continuous rehabilitation training periods are acquired; an initial decision matrix is constructed by using the data of the plurality of periods, the matrix is normalized and weighted, and the positive ideal solution and the negative ideal solution of each index are determined. In the application, a personalized rehabilitation evaluation model based on TOPSIS is used to construct a dynamic evaluation benchmark by using historical data of the user, the closeness degree of the current period and the best / worst state in the history is calculated, an individualized comprehensive evaluation index is generated, and the progress amplitude of the user is truly reflected.

Description

Technical Field

[0001] This invention relates to a rehabilitation motivation and health management system based on cloud-based data visualization, belonging to the field of rehabilitation medical device technology. Background Technology

[0002] Currently, assistive devices for home-based elderly care and community rehabilitation mainly include traditional walking aids, electric wheelchairs, and exoskeleton robots. Traditional walking aids assist users in walking through a support structure and are suitable for people with mild mobility impairments. Electric wheelchairs rely on electric power to achieve mobility and are widely used by people with mobility difficulties. Exoskeleton robots assist limb movement through external power and are mainly used in rehabilitation training scenarios. With the increasing aging of the population, the application of various rehabilitation assistive devices in the fields of home-based elderly care and community rehabilitation is becoming increasingly widespread. Among them, the cloud-based data visualization-based rehabilitation incentive and health management system integrates patient rehabilitation data using cloud computing technology, displays rehabilitation progress through intuitive charts, and combines gamified incentive mechanisms to improve patient compliance. It is an intelligent health platform that collects physiological indicators and exercise data in real time, uses AI algorithms to generate personalized rehabilitation plans, presents health trends in a visual way, and sets incentive elements such as achievement badges and points rankings to enhance patients' rehabilitation motivation. It realizes a visualized rehabilitation incentive and health management system for remote collaborative management between doctors and patients. However, traditional assessment schemes lack real-time collection and comprehensive analysis of users' multi-dimensional physiological and motor data, making it impossible to fully assess rehabilitation effects. Moreover, the evaluation criteria are mostly fixed thresholds or uniform indicators, ignoring individual differences and failing to truly reflect the user's own progress, resulting in limited motivational effects. Therefore, there is an urgent need to improve a rehabilitation motivation and health management system based on cloud data visualization to solve the above-mentioned problems. Summary of the Invention

[0003] The purpose of this invention is to provide a rehabilitation incentive and health management system based on cloud data visualization, in order to solve the problems of traditional assessment schemes lacking real-time collection and comprehensive analysis of users' multi-dimensional physiological and motor data, making it impossible to comprehensively assess rehabilitation effects; and the evaluation standards are mostly fixed thresholds or uniform indicators, ignoring individual differences, making it difficult to truly reflect the user's own progress, and the incentive effect is limited.

[0004] To achieve the above objectives, the present invention provides the following technical solution: A cloud-based data visualization-based rehabilitation incentive and health management system executes a personalized rehabilitation assessment and incentive process based on the TOPSIS model in the following manner during system operation: Step 1: Obtain data on joint motor torque feedback, hub motor acceleration, gait frequency, and bioelectric signal intensity feedback from the user over multiple consecutive rehabilitation training cycles; Step 2: Construct an initial decision matrix using data from the multiple periods, normalize and weight the matrix, and determine the positive and negative ideal solutions for each indicator; Step 3: Calculate the Euclidean distance between the current period data and the positive and negative ideal solutions, and then calculate the relative closeness as a personalized rehabilitation evaluation score for the current period relative to the user's own historical performance. Step 4: Based on the threshold range of the relative closeness, match the preset incentive level and incentive information, and push them to the client interaction platform for visualization. Step 5: The client-side interactive platform displays evaluation scores, historical comparison trends, and personalized incentive feedback through a graphical interface, forming a closed loop of rehabilitation incentives based on the user's own historical dynamic benchmark.

[0005] Furthermore, step 2 specifically includes the following steps: Step 21: Construct the initial decision matrix Where m is the number of cycles and n is the number of evaluation indicators. This represents the value of the j-th indicator in the i-th period; Step 22: Perform vector normalization on the matrix: The normalized matrix is ​​obtained. ; Step 23: Set the weight of each indicator as follows: Construct a weighted norm matrix ,in ; Step 24: Determine the ideal solution and negative ideal solution For positive indicators such as joint motor torque, gait frequency, and bioelectrical signal intensity data, , For negative indicators such as the number of abnormal postures, , ; Step 3 specifically includes the following steps: Step 31: Calculate the distance to the positive ideal solution for each cycle. ; Step 32: Calculate the distance to the negative ideal solution in each period. ; Step 33: Calculate the relative proximity ,in , The closer the value is to 1, the closer the performance of that cycle is to its historical best. The TOPSIS multi-attribute decision model is applied to the evaluation of rehabilitation training effects. Using users' own historical data as the evaluation benchmark, it solves the problem of traditional evaluation methods ignoring individual differences. Normalization eliminates the influence of dimensions, weighting reflects the differences in the importance of indicators, Euclidean distance calculation realizes the comprehensive measurement of multiple indicators, and relative proximity intuitively reflects the gap between the user's performance and their own best level, forming a scientific quantitative evaluation system.

[0006] Furthermore, in step 2, when constructing the initial decision matrix, the number of user historical usage cycles selected is the most recent 7 to 30 days, and abnormal cycle data is removed or smoothed. The abnormal cycle data refers to data missing or deviating beyond a preset threshold due to sensor failure or abnormal user use. The reasonable range of the historical data window (7-30 days) ensures the dynamism of the evaluation (recent data reflects the current state) and avoids evaluation distortion caused by short-term fluctuations. The introduction of the abnormal data processing mechanism effectively eliminates interference factors such as equipment failure or abnormal use, and improves the reliability of the evaluation results.

[0007] Furthermore, the weights of each indicator in step 23 are as follows: The analytic hierarchy process (AHP) is used to determine the judgment matrix, which is constructed based on expert experience.

[0008] Furthermore, in steps 31 and 32, when calculating the Euclidean distance, a weighted Euclidean distance is used. This weighted Euclidean distance is combined with the index weights determined in step S23. The specific calculation formula is as follows: , By incorporating weights into distance calculations, important indicators are given greater weight in the evaluation, and the evaluation results better reflect the core rehabilitation goals.

[0009] Furthermore, the specific method for matching the preset incentive level based on the threshold range of the relative closeness in step 4 is as follows: the relative closeness is divided into four ranges: [0, 0.4), [0.4, 0.6), [0.6, 0.8), and [0.8, 1], which correspond to four incentive levels: "Needs Attention," "Normal Fluctuation," "Stable and Excellent," and "Outstanding Breakthrough." Each level is associated with different visual, auditory, or gamified incentive feedback, transforming the continuous relative closeness score into discrete incentive levels, making it easier for users to understand and receive feedback. The setting of the four levels covers the entire performance range from low to high, which can provide care and reminders when performance is poor, and provide strong positive incentives when breakthroughs are made, forming continuous motivation for recovery.

[0010] Furthermore, it includes a data acquisition unit configured in the intelligent mobile rehabilitation assistive robot, used to collect multimodal physiological and motor data in real time during the user's rehabilitation training process. The multimodal data includes at least joint motor torque feedback, hub motor acceleration, gait frequency and bioelectrical signals. A cloud server, which is communicatively connected to the data acquisition unit, is used to receive and store the multimodal data. The cloud server has a built-in personalized rehabilitation evaluation model based on TOPSIS (Topology for Approximating Ideal Solutions). The model constructs a decision matrix using data from multiple historical usage cycles of the user, determines the ideal solution and the negative ideal solution, calculates the relative closeness of the current cycle data to the historical data, and generates personalized rehabilitation evaluation results. The client-side interaction platform is used to receive and display the rehabilitation evaluation results and corresponding incentive information pushed by the cloud server; A complete closed loop of "data collection - cloud analysis - visualization incentives" has been constructed, which systematizes and automates the originally fragmented rehabilitation training process. By introducing the TOPSIS model, a leap from "uniform standard evaluation" to "individualized dynamic evaluation" has been achieved, so that the evaluation results can more accurately reflect the actual progress of users and avoid the problem of insufficient incentives for some users due to the traditional "one-size-fits-all" standard.

[0011] Compared with the prior art, the beneficial effects of the present invention are: This invention achieves comprehensive perception of the user's rehabilitation status by setting up a multimodal data acquisition unit to collect biomechanical parameters, kinematic parameters, physiological electrical signals, and safety parameters in real time during user rehabilitation training. It employs a TOPSIS-based personalized rehabilitation evaluation model to construct a dynamic evaluation benchmark using the user's historical data, calculating the closeness of the current cycle to the historical best / worst state, and generating an individualized comprehensive evaluation index to accurately reflect the user's progress. Data is remotely stored and analyzed via a cloud server, generating visualized rehabilitation reports and personalized incentive information, which are then pushed to the client interaction platform. The client interaction platform integrates diverse incentive feedback such as progress bars, trend charts, virtual badges, and gamified scenarios, dynamically adjusting the incentive format according to the evaluation level. It supports preprocessing mechanisms such as abnormal cycle data removal and dynamic adjustment of historical windows to ensure stable and reliable evaluation results, thus forming a complete "data acquisition-cloud analysis-visualized incentive" rehabilitation management closed loop, effectively improving user adherence and effectiveness in rehabilitation training. Attached Figure Description

[0012] Figure 1 This is a flowchart of the personalized rehabilitation evaluation and incentive process based on the TOPSIS model of this invention. Detailed Implementation

[0013] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0014] like Figure 1 As shown in the figure, this embodiment provides a cloud-based data visualization-based rehabilitation incentive and health management system, deployed in the home environment of Mr. Wang, a 65-year-old patient with sequelae of stroke; the system includes the following components: Intelligent mobile rehabilitation robot: Equipped with a data acquisition unit, including joint motor torque sensors, wheel hub motor accelerometers, depth cameras, bioelectric sensors (surface electromyography), and posture sensors. The robot body has a touchscreen and is equipped with an interactive terminal developed using QT, supporting voice interaction and touch operation; Cloud server: Deployed on a cloud service platform, running a personalized rehabilitation assessment model based on TOPSIS, configured with a MySQL database to store user data, and communicating with the robot and mobile APP via 4G / Wi-Fi; Mobile App: Installed on the mobile phones of patients' children and the terminals of community rehabilitation doctors, it uses ECharts to visualize data and allows real-time viewing of patients' rehabilitation progress and health reports; Before initial use, the system uses lidar and depth cameras on the robot to scan and model the patient's home environment, generating an indoor map. The patient is then linked to the robot via a secure connection device, and the system records the patient's basic information (age, medical history, rehabilitation goals, etc.) to establish a personal health record.

[0015] Daily rehabilitation training and data collection: Patients undergo 30 minutes of rehabilitation training every morning. The training format includes: Assisted walking mode: The patient holds onto the robotic arms with both hands and performs walking training indoors. The robot assists walking through motors while simultaneously monitoring the patient's balance in real time using posture sensors; Gamified training mode: The robot screen displays virtual scenes (such as the "obstacle course" game), and the patient controls the game character by walking, integrating entertainment elements into the training process; During training, the data acquisition unit collects the following multimodal data in real time at a sampling frequency of 50Hz: joint motor torque feedback (reflecting force level), hub motor acceleration (reflecting gait stability), gait frequency (identified by a depth camera), bioelectrical signals (reflecting muscle activation level), and number of abnormal postures (detecting fall risk events through posture sensors).

[0016] After each training session, the robot filters and removes noise from the raw data, extracts the statistical features (mean, variance, peak value) of the training session, and automatically uploads them to the cloud server via a 4G network.

[0017] Cloud-based personalized rehabilitation assessment: The cloud server receives and stores the training data uploaded by patients daily; when it is necessary to evaluate the rehabilitation effect on a particular day (referred to as the "current period"), the system performs the following steps: Step 1: Construct a historical benchmark dataset The server automatically selects the most recent 7 days (configurable to 7-30 days) before the current date as the historical baseline period. The system performs anomaly detection on historical data: if data for a certain day is missing or deviates beyond a preset threshold due to sensor malfunction or improper use, the period is automatically removed and supplemented forward to ensure that the historical baseline set contains 7 valid periods.

[0018] Step 2: Construct an evaluation index system The system sets four evaluation indicators: average torque of joint motors (positive indicator, the higher the better), average gait frequency (positive indicator, the higher the better), average bioelectric signal (positive indicator, the higher the better), and number of abnormal postures (negative indicator, the lower the better). Step 3: Determine the indicator weights The server uses the entropy weighting method to automatically calculate the weights of each indicator. By analyzing the dispersion of historical data, the indicators that best distinguish differences in user performance (such as the number of abnormal postures) are identified and assigned higher weights. The weight calculation results satisfy the condition that the sum of the weights of all indicators is 1.

[0019] Step 4: Construct the TOPSIS evaluation model The system constructs an initial decision matrix using data from 7 historical periods and the current period, totaling 8 periods, and then performs the following steps: vector normalization to eliminate the influence of dimensions; Construct a weighted normalization matrix by combining the weights; Determine the positive ideal solution (positive index takes the maximum value, negative index takes the minimum value) and the negative ideal solution (positive index takes the minimum value, negative index takes the maximum value). Calculate the weighted Euclidean distance between each cycle and the positive and negative ideal solutions; Calculate the relative closeness (a value between 0 and 1) as a personalized rehabilitation evaluation score for the current period relative to the user's own historical performance; Step 5: Output of evaluation results The closer the relative similarity is to 1, the closer the performance is to the user's best historical state; the closer it is to 0, the closer the performance is to the worst historical state.

[0020] Personalized incentive feedback: The cloud server matches a preset incentive level based on the threshold range of relative relevance.

[0021] Incentive information is pushed to the client interaction platform via the cloud: Robot screen: Displays the daily evaluation score in real time (converted to a percentage or star rating), and shows a comparison with historical periods in the form of progress bars, line graphs, etc. Different visual and auditory feedback is triggered according to the incentive level, such as playing a celebratory animation and voice praise when "outstanding breakthrough" is achieved.

[0022] Mobile App: Generates daily rehabilitation reports with illustrations, including evaluation scores, trend charts, badges, and achievements. It also supports data export and sharing, allowing patients' children or community doctors to view the data at any time.

[0023] Long-term health management: The cloud server continuously accumulates patients' rehabilitation data and generates periodic health reports such as weekly and monthly reports. The report content includes: rehabilitation trend analysis (relative closeness change curve), fluctuation of various indicators, abnormal event records (such as fall risk warnings), and rehabilitation suggestions (such as "recent gait stability has improved, and training time can be appropriately increased"). Patients' families and doctors can remotely monitor the recovery progress through the APP. When a persistently low score or abnormal event occurs, the system will automatically send an early warning notification to enable timely intervention and guidance.

[0024] Through the above specific implementation methods, this system uses the user's own historical data as a benchmark, avoiding the evaluation bias of users with different physical conditions under a uniform standard, truly reflecting individual progress, integrating multiple parameters such as muscle strength, gait, muscle activation, and safety to comprehensively measure rehabilitation effects; from data collection and cloud analysis to visual incentives, a complete rehabilitation motivation cycle is formed, significantly improving user training compliance; at the same time, family members and doctors can understand the patient's status in real time, intervene in a timely manner, and reduce care pressure.

[0025] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0026] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A rehabilitation incentive and health management system based on cloud-based data visualization, characterized in that: During system operation, the personalized rehabilitation assessment and incentive process based on the TOPSIS model is executed in the following manner: Step 1: Obtain data on joint motor torque feedback, hub motor acceleration, gait frequency, and bioelectric signal intensity feedback from the user over multiple consecutive rehabilitation training cycles; Step 2: Construct an initial decision matrix using data from the multiple periods, normalize and weight the matrix, and determine the positive and negative ideal solutions for each indicator; Step 3: Calculate the Euclidean distance between the current period data and the positive and negative ideal solutions, and then calculate the relative closeness as a personalized rehabilitation evaluation score for the current period relative to the user's own historical performance. Step 4: Based on the threshold range of the relative closeness, match the preset incentive level and incentive information, and push them to the client interaction platform for visualization. Step 5: The client-side interactive platform displays evaluation scores, historical comparison trends, and personalized incentive feedback through a graphical interface, forming a closed loop of rehabilitation incentives based on the user's own historical dynamic benchmark.

2. The personalized rehabilitation evaluation and incentive process based on the TOPSIS model according to claim 1, characterized in that: Step 2 specifically includes the following steps: Step 21: Construct the initial decision matrix Where m is the number of cycles and n is the number of evaluation indicators. This represents the value of the j-th indicator in the i-th period; Step 22: Perform vector normalization on the matrix: The normalized matrix is ​​obtained. ; Step 23: Set the weight of each indicator as follows: Construct a weighted norm matrix ,in ; Step 24: Determine the ideal solution and negative ideal solution For positive indicators such as joint motor torque, gait frequency, and bioelectrical signal intensity data, , For negative indicators such as the number of abnormal postures, , .

3. The personalized rehabilitation evaluation and incentive process based on the TOPSIS model according to claim 1, characterized in that: Step 3 specifically includes the following steps: Step 31: Calculate the distance to the positive ideal solution for each cycle. ; Step 32: Calculate the distance to the negative ideal solution in each period. ; Step 33: Calculate the relative proximity ,in , The closer it is to 1, the closer the performance of that cycle is to its historical best.

4. The personalized rehabilitation evaluation and incentive process based on the TOPSIS model according to claim 1, characterized in that: In step 2, when constructing the initial decision matrix, the number of user historical usage cycles selected is the most recent 7 to 30 days, and abnormal cycle data is removed or smoothed. Abnormal cycle data refers to data that is missing or deviates beyond a preset threshold due to sensor failure or user non-normal use.

5. The personalized rehabilitation evaluation and incentive process based on the TOPSIS model according to claim 2, characterized in that: The weights of each indicator in step 23 are as follows: The analytic hierarchy process (AHP) is used to determine the judgment matrix, which is constructed based on expert experience.

6. The personalized rehabilitation evaluation and incentive process based on the TOPSIS model according to claim 3, characterized in that: In steps 31 and 32, the Euclidean distance is calculated using a weighted Euclidean distance, which is combined with the index weights determined in step S23. The specific calculation formula is as follows: , .

7. The personalized rehabilitation evaluation and incentive process based on the TOPSIS model according to claim 1, characterized in that: The specific method for matching the preset incentive level according to the threshold range of the relative proximity in step 4 is as follows: the relative proximity is divided into four ranges: [0,0.4), [0.4,0.6), [0.6,0.8), and [0.8,1], which correspond to four incentive levels: "needs attention", "normal fluctuation", "stable and excellent" and "outstanding breakthrough", respectively. Each level is associated with different visual, auditory or gamified incentive feedback.

8. The rehabilitation incentive and health management system based on cloud data visualization according to claim 1, characterized in that: It includes a data acquisition unit, configured in an intelligent mobile rehabilitation assistive robot, used to collect multimodal physiological and motor data in real time during the user's rehabilitation training. The multimodal data includes at least joint motor torque feedback, hub motor acceleration, gait frequency and bioelectrical signals. A cloud server, which is communicatively connected to the data acquisition unit, is used to receive and store the multimodal data. The cloud server has a built-in personalized rehabilitation evaluation model based on TOPSIS. The model constructs a decision matrix using data from multiple historical usage cycles of the user, determines the ideal solution and the negative ideal solution, calculates the relative closeness of the current cycle data to the historical data, and generates personalized rehabilitation evaluation results. The client-side interaction platform is used to receive and display the rehabilitation evaluation results and corresponding incentive information pushed by the cloud server.

Images