Intelligent generation method and management system for movement scheme of hypertensive patient

Abstract

The invention relates to the technical field of intelligent exercise management, in particular to an intelligent generation method and management system for an exercise scheme of a hypertensive patient, and the method comprises the following steps: firstly, obtaining physical state data of the patient; then, constructing a prediction model; taking the physical state data of the patient as the input of the prediction model, taking the movement effect as the output of the prediction model, and training the prediction model by adopting a ten-fold cross validation method based on an XGBoost algorithm; next, predicting the exercise effect of the patient by adopting the trained model, and generating a personalized exercise scheme; medical assistance is provided according to conditions in the movement process of the patient; finally, the movement condition of the patient is collected, and the movement effect is fed back. According to the method and the management system provided by the invention, effective communication between the patient end and the management end can be realized, personalized exercise guidance is provided to help the hypertensive patient to better manage own exercise, and the exercise scheme is adjusted in real time according to the physical state data of the patient during exercise.

Description

Technical Field

[0001] The present application relates to the technical field of intelligent exercise management, and in particular to a method and management system for intelligently generating exercise plans for patients with hypertension. Background Art

[0002] High blood pressure, a common chronic disease, and the cardiovascular and cerebrovascular diseases it causes have become a major health threat to Chinese residents. According to the World Health Organization, cardiovascular and cerebrovascular diseases are the leading cause of death worldwide, accounting for over 17 million deaths annually. The complications caused by high blood pressure not only impose a heavy medical burden but also place enormous economic pressure on families and society as a whole.

[0003] To effectively address this daunting challenge, the "Healthy China Action Plan (2019-2030)" explicitly calls for integrating exercise management for patients with chronic diseases into cardiovascular and cerebrovascular disease prevention and treatment initiatives. Hypertension, as one of the most significant and modifiable risk factors for cardiovascular and cerebrovascular diseases, is widely considered a key entry point for their prevention and control. Research has shown that early, moderate exercise can regulate blood pressure and enhance cardiovascular health, providing patients with an effective non-drug treatment option.

[0004] Currently, there are several exercise management systems on the market for patients with chronic diseases, which primarily assist patients by assessing their activity levels, changes in vital signs during exercise, and exercise risk. However, these systems generally have some drawbacks:

[0005] (1) Lack of dynamic adjustment mechanism: The existing system is often static and cannot be dynamically adjusted according to the patient's condition after exercise, resulting in the exercise plan being no longer applicable when the condition changes;

[0006] (2) Lack of professional guidance and support: Existing technologies generally lack effective communication channels between patients and medical staff, and patients find it difficult to obtain timely professional advice and guidance when faced with exercise choices;

[0007] (3) Single form of exercise: The existing system's exercise recommendations are limited to a few common aerobic exercises (such as walking and jogging), which lack diversity and may lead to patients' lack of interest in exercise;

[0008] (4) Lack of emergency call function: The existing system does not provide a one-touch emergency call function. When patients have an accident while exercising, they cannot immediately receive professional medical assistance;

[0009] (5) Lack of supervision and incentive mechanisms: Patients lack effective supervision and incentives when implementing exercise plans, which may lead to poor implementation of the exercise plan or abandonment midway;

[0010] (6) Lack of feedback mechanism: The existing system makes it difficult to convey patients’ usage experience and suggestions to developers and medical staff, thus affecting the optimization and improvement of the system. Summary of the Invention

[0011] The present application provides an intelligent generation method and management system for exercise plans for patients with hypertension, which can achieve effective communication between the patient side and the management side, provide personalized exercise guidance to help patients with hypertension better manage their own exercise, and adjust the exercise plan in real time according to the patient's physical condition data during exercise.

[0012] To solve the above technical problems, an embodiment of the present application provides an intelligent generation method and management system for exercise plans for patients with hypertension, which includes the following steps: first, obtaining the patient's physical condition data; then, building a prediction model; using the patient's physical condition data as the input of the prediction model, and the exercise effect as the output of the prediction model, the prediction model is trained using a ten-fold cross-validation method based on the XGBoost algorithm; next, using the trained model to predict the patient's exercise effect, providing the patient with exercise type recommendations, and generating personalized exercise plans based on the different types of exercise selected by the patient; providing medical assistance according to the situation during the patient's exercise; finally, collecting the patient's exercise status and providing feedback on the exercise effect.

[0013] In some exemplary embodiments, the prediction model includes: an exercise pyramid unit, a recommended exercise type module, an inappropriate exercise type module, a recommended exercise timing module and an algorithm module; the exercise pyramid unit is used to divide exercise into different levels according to the number of exercises, exercise intensity and exercise time; the recommended exercise type module is used to generate an exercise type recommendation list based on the patient's personal profile unit and physical condition data module information, and send the exercise type recommendation list to the recommended exercise type module; the inappropriate exercise type module is used to generate an inappropriate exercise list based on the patient's physical condition data module information, and send the inappropriate exercise type list to the inappropriate exercise module; the recommended exercise timing module is used to generate daily recommended exercise timing based on the exercise type recommendation list; the algorithm module generates an exercise plan based on the exercise pyramid, the recommended exercise type module, the inappropriate exercise type module and the recommended exercise timing module.

[0014] In some exemplary embodiments, in a personalized exercise plan, the content of the exercise plan includes: exercise method, exercise intensity, exercise time, exercise frequency, exercise goal, precautions and handling of sudden accidents, and provides video or graphic tutorials to help patients complete the movements correctly.

[0015] In some exemplary embodiments, the physical state data includes resting state data and post-exercise state data; wherein, the resting state data includes the patient's age, gender, height, weight, resting heart rate, resting systolic blood pressure, diastolic blood pressure, BMI index and cardiovascular health indicators; the post-exercise state data is obtained through a sports watch, and the post-exercise state data is refreshed every 5 minutes during the patient's exercise, including the patient's heart rate, systolic blood pressure and diastolic blood pressure after exercise, and heart rate, systolic blood pressure and diastolic blood pressure after exercise recovery.

[0016] In some exemplary embodiments, when using a trained model to predict the patient's exercise effect, the patient's exercise tolerance is evaluated based on the patient's physical condition data to provide the patient with exercise type recommendations; the exercise tolerance is calculated using the following formula:

[0017] RPP=heart rate*systolic blood pressure

[0018] At rest: 6000<RPP<12000

[0019] In motion: 15000<RPP<25000

[0020] Wherein, RPP represents the cardiovascular health index, which is used to reflect the workload of the heart during exercise or stress, and helps to assess the cardiovascular health level;

[0021] Provide medical assistance according to the situation during the patient's exercise, including: if the patient encounters an accident during exercise and needs emergency medical assistance, you can quickly call the emergency number by calling 120 with one click. At the same time, the paramedics can accurately locate the patient's position and provide timely rescue.

[0022] In some exemplary embodiments, a prediction model is trained using a ten-fold cross-validation method based on the XGBoost algorithm, including: generating a set of classification trees using the XGBoost algorithm, training the prediction model using the ten-fold cross-validation method for each classification tree, adding the scores of the leaf nodes corresponding to each tree, calculating a prediction estimate for the features of the i-th sample, establishing an objective function, and then defining the prediction estimate output by the machine learning model as the presence or absence of an exercise program effect after rounding; wherein i is a natural number from 1 to n, n represents the number of samples of patients with metabolic diseases used for training, and n is a natural number greater than or equal to 1000.

[0023] In some exemplary embodiments, the calculation formula for the predicted estimated value calculated for the feature of the i-th sample is shown in Formula 1:

[0024]

[0025] Among them, f kRepresents the prediction results of each regression tree, K is the number of regression trees, x i Represents the characteristics of the i-th sample, y i Represents the predicted estimated value of the i-th sample. Formula 1 indicates that given an input value x i , the output value is the predicted estimated value y of K regression trees i .

[0026] On the second aspect, the embodiment of the present application also provides an intelligent generation and management system for exercise plans for patients with hypertension, which generates an exercise guide based on the intelligent generation method for exercise plans for patients with hypertension described in the above embodiment, including: a patient end and a management end, and the patient end is communicatively connected to the management end; wherein the patient end includes: a personal center module, a body state data module, an exercise guide module, a one-button call module and an exercise feedback module; the personal center module is used to store patient personal information; the body state data module is used to evaluate the patient's exercise risk based on the patient's personal information and obtain the patient's body state data; the exercise guide module is used to predict the patient's exercise effect, provide the patient with exercise type recommendations, and generate personalized exercise plans based on the different exercise types selected by the patient; the one-button call module is used to call 120 with one button when an accident occurs during the patient's exercise and emergency medical assistance is required, so that the emergency number can be quickly dialed, and the rescue personnel can accurately locate the patient's position and provide timely rescue; the exercise feedback module is used to display the patient's exercise status and provide feedback on the exercise effect.

[0027] In some exemplary embodiments, the exercise feedback module includes: an exercise note module, which is used to record the patient's daily exercise method, intensity and amount of exercise, and whether there is any discomfort after exercise; a sports ambassador honor list unit, which is used to sort the patients according to the frequency of use of the system, so as to reward patients with high frequency of use according to the ranking; a feedback unit, which is used to collect patients' opinions on the use experience, function suggestions and exercise recommendations of the system; an online consultation unit, which is used to provide medical staff with an online consultation channel so that patients can consult online about questions about exercise plan selection; an exit unit, which is used to exit the current patient login information; and a system version viewing unit, which is used to view the current system version information.

[0028] In some exemplary embodiments, the management end includes: an exercise update module for updating the exercise content of the system; an information reading module for viewing patients' questions about exercise plan selection so that medical staff can respond; an information export module for exporting information from the patient end; and a permission setting module for setting the use permissions of the system.

[0029] The technical solution provided by the embodiments of the present application has at least the following advantages:

[0030] The embodiment of the present application provides an intelligent generation method and management system for exercise plans for patients with hypertension, which includes the following steps: first, obtaining the patient's physical condition data; then, building a prediction model; using the patient's physical condition data as the input of the prediction model and the exercise effect as the output of the prediction model, the prediction model is trained using a ten-fold cross-validation method based on the XGBoost algorithm; next, using the trained model to predict the patient's exercise effect, providing the patient with exercise type recommendations, and generating personalized exercise plans based on the different types of exercise selected by the patient; providing medical assistance according to the situation during the patient's exercise; finally, collecting the patient's exercise status and providing feedback on the exercise effect. The present application constructs a prediction model based on the XGBoost machine learning algorithm, which can provide patients with tailored personalized exercise guides, assist doctors in forming more effective exercise plans, and improve the effect of exercise intervention.

[0031] The embodiments of the present application provide an intelligent method for generating exercise plans and a management system for patients with hypertension. On the one hand, the intelligent method for generating exercise plans and the management system of the present application can obtain the patient's real-time health data through the smart watch worn by the patient, and dynamically adjust the activity plan according to the health data, thereby reducing health risks. On the other hand, the intelligent generation method and management system of the exercise plan of the present application improves information interactivity; the system of the present application has a real-time communication function, and patients can consult medical staff about exercise-related issues at any time, and medical staff can also provide feedback and guidance in time to improve the exercise effect; in addition, the exercise recommended by the system of the present application includes but is not limited to several common aerobic exercises (such as walking, jogging), and also provides strength training, flexibility training, yoga, etc. According to the different types of exercise selected by the patient, a personalized exercise plan is generated, and a video or graphic tutorial is provided to help the patient complete the action correctly. If medical assistance is needed during exercise, 120 can be called with one click, and the rescue personnel can accurately locate the patient's position and provide timely rescue; moreover, the system of the present application is also equipped with a supervision and incentive mechanism and an intelligent feedback mechanism. By setting up a sports ambassador honor list unit, patients who rank high on the list are rewarded to encourage them to persist in exercise; and by collecting patient feedback, the system can continuously optimize the exercise plan, gradually form the best exercise plan suitable for the patient, and improve the overall user experience. Finally, the system of the present application also has the advantage of multi-platform support. The system of the present application can be used on a variety of devices, such as mobile applications, web pages, applets, etc., making it convenient for patients to manage their exercise anytime, anywhere. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] One or more embodiments are exemplarily described by pictures in the corresponding drawings. These exemplifications do not constitute limitations on the embodiments. Unless otherwise stated, the pictures in the drawings do not constitute proportional limitations.

[0033] Figure 1 A flowchart of a method for intelligently generating an exercise plan for hypertensive patients provided in an embodiment of the present application.

[0034] Figure 2 This is a structural block diagram of an intelligent generation and management system for exercise plans for hypertensive patients provided in an embodiment of the present application.

[0035] Figure 3 This is a diagram of the login interface of the login module in the intelligent generation and management system for exercise plans for hypertensive patients provided in an embodiment of the present application.

[0036] Figure 4 This is a structural diagram of the patient side of the intelligent generation and management system for exercise plans for hypertensive patients provided in an embodiment of the present application.

[0037] Figure 5 The data input interface of the data input module in the intelligent generation and management system for exercise programs for hypertensive patients provided in the embodiment of the present application Figure 1 .

[0038] Figure 6 The data input interface of the data input module in the intelligent generation and management system for exercise programs for hypertensive patients provided in the embodiment of the present application Figure 2 .

[0039] Figure 7 This is a diagram of the exercise guide interface for the intelligent generation and management system for exercise plans for hypertensive patients provided in an embodiment of the present application.

[0040] Figure 8 This is a structural diagram of the management end of the intelligent generation and management system for exercise plans for hypertensive patients provided in an embodiment of the present application. DETAILED DESCRIPTION

[0041] As can be seen from the background technology, the existing exercise management systems for patients with chronic diseases mainly assist patients by evaluating their own activity levels, changes in vital signs during exercise, and exercise risks. However, there are technical problems such as lack of dynamic adjustment mechanism, lack of professional guidance and support, single exercise form, lack of emergency call function, lack of supervision and incentive mechanism, and lack of feedback mechanism.

[0042] In order to solve the above technical problems, the embodiment of the present application provides an intelligent generation method and management system for exercise plans for patients with hypertension, which method includes the following steps: first, obtaining the patient's physical condition data; then, building a prediction model; using the patient's physical condition data as the input of the prediction model, using the exercise effect as the output of the prediction model, and training the prediction model using the ten-fold cross-validation method based on the XGBoost algorithm; next, using the trained model to predict the patient's exercise effect, providing the patient with a recommendation for the type of exercise, generating a personalized exercise plan based on the patient's selected type of exercise, and providing medical assistance according to the situation during the patient's exercise; finally, collecting the patient's exercise status and providing feedback on the exercise effect. The present application provides an intelligent generation method and management system for exercise plans for patients with hypertension, which can realize effective communication between the patient end and the management end, provide personalized exercise guidance to help hypertensive patients better manage their own exercise, and adjust the exercise plan in real time based on the patient's physical condition data during exercise.

[0043] The following detailed description of the various embodiments of the present application is provided in conjunction with the accompanying drawings. However, those skilled in the art will appreciate that many technical details are provided in the various embodiments of the present application to facilitate a better understanding of the present application. However, even without these technical details and the various variations and modifications based on the following embodiments, the technical solutions claimed in the present application can still be implemented.

[0044] See Figure 1 The embodiment of the present application provides a method and management system for intelligently generating an exercise plan for a patient with hypertension, the method comprising the following steps:

[0045] Step S1: Acquire patient's physical condition data.

[0046] Step S2: Construct a prediction model; use the patient's physical condition data as the input of the prediction model, use the exercise effect as the output of the prediction model, and train the prediction model using the ten-fold cross-validation method based on the XGBoost algorithm.

[0047] Step S3: Use the trained model to predict the patient's exercise effect, provide the patient with exercise type recommendations, and generate a personalized exercise plan based on the patient's selected exercise type.

[0048] Step S4: provide medical assistance according to the situation while the patient is exercising.

[0049] Step S5: Collect the patient's exercise status and provide exercise effect feedback.

[0050] In some embodiments, the physical state data in step S1 includes resting state data and post-exercise state data; wherein, the resting state data includes the patient's age, gender, height, weight, resting heart rate, resting systolic blood pressure, diastolic blood pressure, BMI index and cardiovascular health indicators; the post-exercise state data is obtained through a sports watch, and the post-exercise state data is refreshed every 5 minutes during the patient's exercise, including the patient's heart rate, systolic blood pressure and diastolic blood pressure after exercise, and heart rate, systolic blood pressure and diastolic blood pressure after exercise recovery.

[0051] In some embodiments, the prediction model in step S2 includes: an exercise pyramid unit, a recommended exercise type module, an inappropriate exercise module, a recommended exercise timing module and an algorithm module; the exercise pyramid unit is used to divide exercise into different levels according to the number of exercises, exercise intensity and exercise time; the recommended exercise type module is used to generate an exercise type recommendation list based on the patient's personal data unit and physical condition data module information, and send the exercise type recommendation list to the recommended exercise type module; the inappropriate exercise module is used to generate an inappropriate exercise list based on the patient's physical condition data module information, and send the inappropriate exercise type list to the inappropriate exercise module; the recommended exercise timing module is used to generate daily recommended exercise timing according to the exercise type recommendation list; the algorithm module generates an exercise plan based on the exercise pyramid, the recommended exercise type module, the inappropriate exercise module and the recommended exercise timing module.

[0052] In some embodiments, in the generated personalized exercise plan, the content of the exercise plan includes: exercise method, exercise intensity, exercise time, exercise frequency, exercise goal, precautions and handling of sudden accidents, and provides video or graphic tutorials to help patients complete the movements correctly.

[0053] It should be noted that when building the prediction model in step S2, the four training parameters of XGBoost are first set.

[0054] The maximum depth of the tree is set to 6. The maximum depth of the tree is the number of tree layers. Generally, the deeper the tree layers, the more complex the model.

[0055] The learning rate is set to 0.3. The learning rate is the distance required to adjust the parameters each time. An appropriate learning rate can accurately find the minimum value of the loss function, which is the optimal solution for the model parameters.

[0056] The motion plan, that is, the output of the prediction model, needs to determine whether the motion plan is valid or invalid. It is a binary classification problem. Therefore, a valid motion plan is defined as 1 and an invalid motion plan is defined as 0.

[0057] The number of iterations indicates how many trees will be used to define the model. Set the number of iterations to 50.

[0058] To set the training parameters in XGBoost, enter the following code:

[0059] param=['max_depth':6,'eta':0.3,'silent':0,'obejctive':'binary:logistic']

[0060] Among them, max_depth represents the maximum depth of the tree, eta represents the learning rate, silent represents the silent mode, objective represents the learning target parameter, and binary:logistic represents the probability of outputting binary classification logistic regression in the model.

[0061] When training the prediction model, the patient's physical condition data is used as the input of the prediction model, and the exercise effect is used as the output of the prediction model. The prediction model is trained using the ten-fold cross-validation method based on the XGBoost algorithm. In the XGBoost machine learning system, call the train function in XGBoost and enter the following code:

[0062] bst=xgb.train(param,dtrain,50)

[0063] Among them, bst represents the result after the final train training, and 50 represents the number of iterations.

[0064] During the call of the train function, calculations are performed according to the following principles:

[0065] The XGBoost algorithm is used to generate a set of classification trees. Each classification tree is trained using the ten-fold cross-validation method. The scores of the leaf nodes corresponding to each tree are added together, and the predicted estimate value of the features of the i-th sample is calculated. The objective function is established, and the predicted estimate value output by the machine learning model is rounded off and defined as the presence or absence of an exercise plan, where i is a natural number from 1 to n.

[0066] Specifically, in some embodiments, the prediction model is trained based on the XGBoost algorithm using a ten-fold cross-validation method, including: using the XGBoost algorithm to generate a set of classification trees, each classification tree is trained using the ten-fold cross-validation method to train the prediction model, adding the scores of the leaf nodes corresponding to each tree, calculating the predicted estimate for the features of the i-th sample, establishing an objective function, and then defining the predicted estimate output by the machine learning model as the presence or absence of the effect of the exercise program after rounding; wherein, i is a natural number from 1 to n, n represents the number of samples of patients with metabolic diseases used for training, and n is a natural number greater than or equal to 1000.

[0067] In some embodiments, the calculation formula for the predicted estimated value calculated for the feature of the i-th sample is shown in Formula 1:

[0068]

[0069] Among them, f k Represents the prediction results of each regression tree, K is the number of regression trees, x i Represents the characteristics of the i-th sample, y i Represents the predicted estimated value of the i-th sample. Formula 1 indicates that given an input value x i , the output value is the predicted estimated value y of K regression trees i .

[0070] In some embodiments, when using a trained model to predict the patient's exercise effect, the patient's exercise tolerance is assessed based on the patient's physical condition data to provide the patient with exercise type recommendations; the exercise tolerance is calculated using the following formula:

[0071] RPP=heart rate*systolic blood pressure

[0072] At rest: 6000<RPP<12000

[0073] In motion: 15000<RPP<25000

[0074] Wherein, RPP represents the cardiovascular health index, which is used to reflect the workload of the heart during exercise or stress, and helps to assess the cardiovascular health level;

[0075] Provide medical assistance according to the situation during the patient's exercise, including: if the patient encounters an accident during exercise and needs emergency medical assistance, you can quickly call the emergency number by calling 120 with one click. At the same time, the paramedics can accurately locate the patient's position and provide timely rescue.

[0076] It's important to note that higher RPP values ​​generally indicate a greater workload on the heart, potentially increasing the risk of cardiovascular events, while lower RPP values ​​indicate a lighter workload and a healthier cardiovascular system. Recording data from multiple time periods, including resting, during exercise, and post-exercise recovery, allows for comprehensive analysis for a more accurate assessment.

[0077] See Figure 2, the embodiment of the present application also provides an intelligent generation and management system for exercise plans for patients with hypertension, which generates an exercise guide based on the intelligent generation method for exercise plans for patients with hypertension described in the above embodiment, including: a patient end and a management end, wherein the patient end is communicatively connected to the management end; wherein the patient end includes: a personal center module 100, a physical state data module 200, an exercise guide module 300, an exercise feedback module 400 and a one-button call module 500; the personal center module 100 is used to store patient personal information; the physical state data module 200 is used to assess the patient's exercise risk based on the patient's personal information and obtain the patient's physical state data; the exercise guide module 300 is used to predict the patient's exercise effect, provide the patient with exercise type recommendations, and generate a personalized exercise plan based on the different exercise types selected by the patient; the exercise feedback module 400 is used to display the patient's exercise status and provide exercise effect feedback; the one-button call module 500 is used to quickly call 120 by one-button calling when an accident occurs during the patient's exercise and emergency medical assistance is required. At the same time, the paramedics can accurately locate the patient's position and provide timely rescue.

[0078] In some embodiments, the exercise feedback module 400 includes: an exercise note module, which is used to record the patient's daily exercise method, intensity and amount of exercise, and whether there is any discomfort after exercise; a sports ambassador honor list unit, which is used to sort the patients according to the frequency of use of the system, so as to reward patients with high frequency of use according to the sorting; a feedback unit, which is used to collect patients' opinions on the use experience, function suggestions and exercise recommendations of the system; an online consultation unit, which is used to provide medical staff with an online consultation channel so that patients can consult online about questions about exercise plan selection; an exit unit, which is used to exit the current patient login information; and a system version viewing unit, which is used to view the current system version information.

[0079] In some embodiments, the management end includes: an exercise update module 600, which is used to update the exercise content of the system; an information reading module 700, which is used to view patients' questions about exercise plan selection so that medical staff can respond; an information export module 800, which is used to export information from the patient end; and a permission setting module 900, which is used to set the usage permissions of the system.

[0080] The system provided in this application provides personalized exercise guidance to help hypertensive patients better manage their own exercise through effective communication between the patient side and the management side, and adjusts the exercise plan in real time based on the patient's physical condition data during exercise. The exercise list includes recommended exercises and exercises that should not be performed. The exercise feedback module enables patients to obtain exercise guidance and support more conveniently through the exercise pyramid, the sports ambassador honor list incentive mechanism and the online consultation function, thereby improving the quality of life and disease management effects. In addition, the system has recording and feedback functions, which support patients to record their exercise conditions and discomfort reactions and allow medical staff to provide patients with one-on-one exercise suggestions.

[0081] Specifically, the intelligent generation method and management system of exercise plans for hypertensive patients provided by this application are introduced in detail through specific embodiments below.

[0082] See also Figures 1 to 8 The intelligent generation method and management system of exercise plans for hypertensive patients in the embodiment of the present application include a patient end and a management end, and the patient end is communicatively connected to the management end.

[0083] In the examples, see Figure 2 As shown, in an embodiment of the present application, a block diagram of a method and management system for intelligently generating an exercise plan for a patient with hypertension is also provided. The system includes a patient-side module (a personal center module 100, a physical status data module 200, an exercise guide module 300, an exercise feedback module 400, and a one-touch call module 500); and a management-side module (an exercise update module 600, an information reading module 700, an information export module 800, and a permission setting module 900).

[0084] See also Figure 3 As shown, the login interface is used for users to log in to the system.

[0085] In an embodiment of the present application, users of the present application system can log in through an account password or a face recognition system.

[0086] See also Figures 4 to 8 As shown, the personal center module 100 is used to edit the patient's personal information, which includes name, ID number, contact number, medical history, and exercise habits. The editing operation includes at least one of adding, modifying, and deleting.

[0087] The physical status data module 200 is used to assess patient exercise risk. Physical status data includes resting state data and post-exercise state data. Post-exercise data is primarily acquired via a sports watch. The patient's exercise tolerance is assessed based on the physical status data. The module determines whether the patient should proceed to the next step based on their health status. If the patient is unfit for exercise, the module prohibits them from proceeding to the next step to ensure patient safety.

[0088] In the present application, the resting state data includes the patient's age, gender, height, weight, resting heart rate, resting systolic blood pressure, diastolic blood pressure, BMI index and cardiovascular health index (RPP).

[0089] In this application, the post-exercise status data refers to the data collected every 5 minutes during the patient's exercise, including the patient's heart rate, systolic blood pressure and diastolic blood pressure after exercise, and heart rate, systolic blood pressure and diastolic blood pressure after exercise recovery.

[0090] It should be noted that the cardiovascular health index (RPP) is calculated using the following formula: RPP = heart rate * systolic blood pressure.

[0091] It should be noted that the cardiovascular health indicator is between 6,000 and 12,000 at rest, indicating normal cardiovascular function. During exercise, it is between 15,000 and 25,000. If the RPP value exceeds 25,000 during exercise, especially if it lasts for a long time or is for patients with cardiovascular disease, it may indicate excessive cardiac stress and a certain risk. A lower RPP value indicates less cardiac stress and a healthier cardiovascular system.

[0092] The system's subsequent operational process is as follows: After entering personal data and physical condition data, an exercise guide is generated. The exercise guide recommends corresponding exercise methods. When patients choose different exercise methods, they will be given suggestions on exercise intensity, exercise time, exercise frequency, exercise goals, etc. Corresponding exercise video tutorials are also available for viewing.

[0093] Based on the above technical solution, the embodiment of the present application provides an intelligent generation method and management system for exercise plans for patients with hypertension, which includes the following steps: first, obtaining the patient's physical condition data; then, building a prediction model; using the patient's physical condition data as the input of the prediction model and the exercise effect as the output of the prediction model, the prediction model is trained using a ten-fold cross-validation method based on the XGBoost algorithm; next, using the trained model to predict the patient's exercise effect, providing the patient with exercise type recommendations, and generating personalized exercise plans based on the different types of exercise selected by the patient; providing medical assistance according to the situation during the patient's exercise; finally, collecting the patient's exercise status and providing feedback on the exercise effect. This application constructs a prediction model based on the XGBoost machine learning algorithm, which can provide patients with customized personalized exercise guides, assist doctors in forming more effective exercise plans, and improve the effect of exercise intervention.

[0094] The present application provides a method and management system for intelligently generating an exercise plan for patients with hypertension, which has the following advantages:

[0095] (1) The system of this application can obtain the patient's real-time health data through the smart watch worn by the patient, and dynamically adjust the activity plan based on the health data, thereby reducing health risks;

[0096] (2) Information interactivity: This application system has a real-time communication function. Patients can consult medical staff about exercise-related issues at any time, and medical staff can also provide timely feedback and guidance to improve exercise effects;

[0097] (3) Rich variety and online tutorials: The exercises recommended by this application system include but are not limited to several common aerobic exercises (such as walking and jogging). It also provides strength training, flexibility training, yoga, etc., and provides video or graphic tutorials to help patients complete the movements correctly;

[0098] (4) Supervision and incentive mechanism: By setting up a sports ambassador honor list, patients who rank high on the list will be rewarded to encourage them to persist in exercising;

[0099] (5) One-touch emergency call function: With one-touch emergency call, rescue procedures can be quickly initiated, greatly reducing response time compared to traditional dialing methods. The one-touch call system integrates positioning technology and can automatically send location information when the patient calls, allowing rescue personnel to arrive at the scene accurately and quickly.

[0100] (6) Intelligent feedback mechanism: By collecting patient feedback, the system can continuously optimize the exercise plan, gradually form the best exercise plan suitable for the patient, and improve the overall user experience.

[0101] (7) Multi-platform support: This application system can be used on a variety of devices, such as mobile applications, web pages, mini-programs, etc., making it convenient for patients to manage their exercise anytime and anywhere.

[0102] Those skilled in the art will appreciate that the above-described embodiments are specific examples for implementing the present application, and that in actual applications, various changes in form and detail may be made thereto without departing from the spirit and scope of the present application. Any person skilled in the art may make changes and modifications without departing from the spirit and scope of the present application. Therefore, the scope of protection of the present application shall be subject to the scope defined in the claims.

Claims

1. A method for intelligently generating an exercise program for patients with hypertension, characterized in that: The following steps are involved: Obtain patient physical status data; Construct a prediction model; use the patient's physical condition data as the input of the prediction model, use the exercise effect as the output of the prediction model, and use the ten-fold cross-validation method based on the XGBoost algorithm to train the prediction model; Use the trained model to predict the patient's exercise effect, provide patients with exercise type recommendations, and generate personalized exercise plans based on the different types of exercise selected by the patient; Provide medical assistance as appropriate during patient movement; Collect the patient's exercise status and provide feedback on the exercise effect.

2. The method for intelligently generating an exercise program for patients with hypertension according to claim 1, characterized in that: The prediction model includes: a sports pyramid unit, a recommended sports type module, an unsuitable sports module, a recommended sports timing module and an algorithm module; The exercise pyramid unit is used to divide exercise into different levels according to the number of exercises, exercise intensity and exercise time; The recommended exercise type module is used to generate an exercise type recommendation list according to the patient's personal data unit and physical condition data module information, and send the exercise type recommendation list to the recommended exercise type module; The inappropriate exercise module is used to generate an inappropriate exercise list according to the patient's physical condition data module information, and send the inappropriate exercise category list to the inappropriate exercise module; The recommended exercise timing module is used to generate daily recommended exercise timings according to the recommended exercise type list; The algorithm module generates an exercise plan according to the exercise pyramid, the recommended exercise type module, the unsuitable exercise type module, and the recommended exercise timing module.

3. The method for intelligently generating an exercise program for patients with hypertension according to claim 1, characterized in that: In the personalized exercise plan, the content of the exercise plan includes: exercise method, exercise intensity, exercise time, exercise frequency, exercise goal, precautions and handling of sudden accidents, and provides video or graphic tutorials to help patients complete the movements correctly.

4. The method for intelligently generating an exercise program for patients with hypertension according to claim 1, characterized in that: The body state data includes resting state data and post-exercise state data; wherein, The resting state data include the patient's age, gender, height, weight, resting heart rate, resting systolic blood pressure, diastolic blood pressure, BMI index and cardiovascular health index; The post-exercise status data is obtained through a sports watch and is refreshed every 5 minutes during the patient's exercise, including the patient's heart rate, systolic blood pressure and diastolic blood pressure after exercise, and heart rate, systolic blood pressure and diastolic blood pressure after exercise recovery.

5. The method for intelligently generating an exercise program for patients with hypertension according to claim 1, characterized in that: When using the trained model to predict the patient's exercise effect, the exercise tolerance is evaluated based on the patient's physical condition data to provide the patient with exercise type recommendations; the exercise tolerance is calculated using the following formula: RPP=heart rate*systolic blood pressure At rest: 6000 <RPP <12000 In motion: 15000 <RPP <25000 In the formula, RPP represents a cardiovascular health index, which is used to reflect the workload of the heart under exercise or stress and helps assess the cardiovascular health level; Provide medical assistance according to the situation during the patient's exercise, including: if the patient encounters an accident during exercise and needs emergency medical assistance, you can quickly dial the emergency number by calling 120 with one click. At the same time, the rescue personnel can accurately locate the patient's position and provide timely rescue.

6. The method for intelligently generating an exercise program for patients with hypertension according to claim 1, characterized in that: The prediction model is trained using the ten-fold cross validation method based on the XGBoost algorithm, including: The XGBoost algorithm was used to generate a set of classification trees. Each classification tree was trained using the ten-fold cross-validation method. The scores of the leaf nodes corresponding to each tree were added together, and the predicted estimate was calculated for the features of the i-th sample. The objective function was established, and the predicted estimate output by the machine learning model was defined as the presence or absence of the effect of the exercise program after rounding. Among them, i is a natural number from 1 to n, n represents the number of samples of patients with metabolic diseases used for training, and n is a natural number greater than or equal to 1000.

7. The method for intelligently generating an exercise program for patients with hypertension according to claim 6, characterized in that: The calculation formula for the predicted estimated value calculated for the features of the i-th sample is shown in Formula 1: Among them, f k represents the prediction result of each regression tree, K is the number of regression trees, x i represents the characteristics of the i-th sample, y i Represents the predicted estimated value of the i-th sample. Formula 1 indicates that given an input value x i , the output value is the predicted estimated value y of K regression trees i .

8. An intelligent generation and management system for exercise plans for patients with hypertension, which generates exercise guidelines based on the intelligent generation method for exercise plans for patients with hypertension as described in any one of claims 1 to 7, characterized in that: include: A patient end and a management end, wherein the patient end is in communication connection with the management end; wherein, The patient terminal includes: a personal center module, a physical status data module, an exercise guide module, a one-key call module and an exercise feedback module; The personal center module is used to store patient personal information; The body status data module is used to evaluate the patient's exercise risk based on the patient's personal information to obtain the patient's body status data; The exercise guidance module is used to predict the patient's exercise effect, provide the patient with exercise type recommendations, and generate a personalized exercise plan based on the different exercise types selected by the patient; The one-key call module is used to quickly dial the emergency number by calling 120 with one key when an accident occurs during the patient's exercise and emergency medical assistance is needed. At the same time, the rescue personnel can accurately locate the patient's position and provide timely rescue; The movement feedback module is used to display the patient's movement status and provide feedback on the movement effect.

9. The intelligent generation and management system for exercise programs for patients with hypertension according to claim 8, characterized in that: The motion feedback module comprises: The exercise note module is used to record the patient's daily exercise method, intensity and amount, as well as whether there is any discomfort after exercise; A sports ambassador honor list unit is used to rank the patients according to the frequency of use of the system, so as to reward the patients with high frequency of use according to the ranking; A feedback unit for collecting patients' opinions on the use experience, function suggestions and exercise recommendations of the system; Online consultation unit, used to provide medical staff with online consultation channels so that patients can consult online about questions about exercise program selection; Logout unit, used to log out the current patient login information; The system version viewing unit is used to view the current system version information.

10. The intelligent generation and management system for exercise programs for patients with hypertension according to claim 8, characterized in that: The management terminal includes: A motion update module, used for updating the motion content of the system; The information reading module is used to view patients' questions about exercise program selection so that medical staff can respond; An information export module, used for exporting the patient's information; The permission setting module is used to set the usage permission of the system.

Images