Cardiac rehabilitation system, method and device for acute myocardial infarction patient and storage medium

Abstract

The invention relates to the field of medical rehabilitation, and discloses a heart rehabilitation system, method and device for an acute myocardial infarction patient and a storage medium, and the system comprises a data processing module which is used for collecting and storing static data and dynamic data of the patient; the exercise optimization module establishes a health target model based on the patient data and generates a personalized exercise prescription; the action monitoring module captures actions of the patient when the patient executes the exercise prescription, generates a skeleton model, carries out comparison based on a standard action template, and evaluates the action quality; the feedback adjustment module dynamically adjusts the exercise prescription according to the training effect and the action deviation of the patient; the user interaction module provides exercise prescription display, real-time feedback and action correction suggestions through the patient terminal, and supports patient file management and rehabilitation plan adjustment through the medical care terminal. Through cooperation of multiple modules, a complete closed-loop rehabilitation management process is formed, the rehabilitation effect and safety are improved, and the system is suitable for a home rehabilitation environment.

Description

Technical Field

[0001] The present invention relates to the technical field of medical rehabilitation, and specifically to a cardiac rehabilitation system, method, device and storage medium for patients with acute myocardial infarction. Background Art

[0002] Cardiac rehabilitation is an important part in the treatment process of patients with acute myocardial infarction (AMI), and plays a significant role in improving the cardiac function of patients, enhancing the quality of life and reducing the readmission rate. However, there are still many deficiencies in the existing cardiac rehabilitation means when implemented in a home environment. Traditional rehabilitation methods usually rely on paper manuals or simple telephone consultations, lacking real-time monitoring of the patient training process and personalized guidance, resulting in difficult-to-guarantee rehabilitation effects. Patients often cannot accurately understand the content of the exercise prescription and the standard requirements for action execution, which makes the exercise effect lower than expected, and may even cause health risks due to improper actions.

[0003] In recent years, with the development of mobile Internet technology and wearable devices, remote rehabilitation platforms have gradually emerged, but these platforms still have problems of low intelligent level and insufficient personalization ability. Most of the exercise prescriptions provided by remote rehabilitation platforms are mainly fixed templates, lacking the ability to dynamically generate based on individual patient data. In addition, the technology in action monitoring is relatively backward. Existing systems usually can only capture simple movement trajectories, and cannot accurately evaluate and correct the action quality of patients, so that patients may perform wrong actions during training, affecting the rehabilitation effect. For key health indicators such as the real-time heart rate and energy consumption of patients during the training process, existing platforms also lack sufficient dynamic feedback and prescription adjustment mechanisms.

[0004] More importantly, the compliance of patients during rehabilitation training in a home environment is often low, and existing systems lack an effective reminder mechanism for the patient rehabilitation plan and a visual display of the training completion rate, making it difficult to motivate patients to continuously participate in training. At the same time, when medical staff remotely manage patients, they face the problems of fragmented data and inability to efficiently master the patient training status, further reducing the efficiency and effect of rehabilitation management. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the present invention provides a cardiac rehabilitation system, method, device and storage medium for patients with acute myocardial infarction, and solves the problems of insufficient intelligence, weak personalization ability, low action monitoring accuracy and lack of dynamic feedback in the existing cardiac rehabilitation system.

[0006] To achieve the above objectives, the present invention is realized through the following technical solutions: A cardiac rehabilitation system for patients with acute myocardial infarction, including: A data processing module, which is used to collect and store the static and dynamic data of patients, including the basic information of patients, cardiopulmonary exercise function parameters, real-time heart rate, blood pressure and energy consumption; A motion optimization module, which is used to generate a personalized exercise prescription for patients based on the static and dynamic data of patients, including exercise type, intensity, time and frequency; An action monitoring module, which is used to monitor the actions of patients performing the exercise prescription, evaluate the action quality of patients based on the deviation between the patient's actions and the standard action template, and provide corrective suggestions; A feedback adjustment module, which is used to dynamically adjust the exercise prescription according to the training effect and action deviation of patients; A user interaction module, which is used to provide exercise prescriptions and feedback information for patients, and to allow medical staff to query patient files and adjust rehabilitation plans.

[0007] Preferably, the motion optimization module includes: A health modeling module, which is used to establish a patient health goal model according to the cardiopulmonary function parameters, muscle strength parameters and balance ability parameters of patients; An optimization solution module, which is used to generate an optimal exercise prescription based on the patient health goal model through an optimization method; A constraint condition module, which is used to set constraints on exercise intensity, heart rate range, joint range of motion and cardiovascular load to ensure the safety of the exercise prescription.

[0008] Preferably, the motion optimization module adopts an optimization method based on the variational method to generate an exercise prescription; wherein, the health goal model is constructed according to the cardiopulmonary function, muscle strength and balance ability of patients, and specifically includes the following goals: By optimizing the cardiopulmonary function parameters of patients, improve exercise endurance; According to the muscle strength parameters and activity ability of patients, adjust the exercise intensity; Improve joint mobility and balance stability through balance training.

[0009] Preferably, the action monitoring module includes: An action capture module, which is used to capture the motion trajectory of patients in real time through a camera or wearable device, and generate a skeletal key point model; A standard comparison module, which is used to match the patient skeletal key point model with the standard action template; A deviation detection module, which is used to calculate the deviation value between the patient's actions and the standard actions based on the deviation measurement method, and trigger corrective suggestions.

[0010] Preferably, the action monitoring module evaluates the action quality based on a graph convolutional network, including the following steps: Generate a skeletal key point map of the patient, where the key points include the spatial coordinates of the shoulders, knees, and joints; Input the patient's skeletal key point map into a graph convolutional network and compare it with a standard action template; Determine the correctness of the patient's action according to the deviation threshold and generate a correction suggestion.

[0011] Preferably, the feedback adjustment module includes: A deviation analysis module for analyzing the difference between the patient's action deviation and the target health parameters; A prescription adjustment module for dynamically adjusting the exercise type, exercise intensity, and exercise time according to the analysis results.

[0012] Preferably, the user interaction module includes: A patient terminal for presenting the rehabilitation plan and feedback information to the patient, including action demonstrations, training completion rates, and correction suggestions; a management terminal for medical staff to query patient data, view the rehabilitation progress, and adjust the exercise prescription.

[0013] The present invention also provides a cardiac rehabilitation method for patients with acute myocardial infarction, including the following steps: Collect the static data and dynamic data of the patient, including the patient's basic information, cardiopulmonary function parameters, and real-time training indicators; Establish a health target model based on the patient's data and generate a personalized exercise prescription through an optimization method; Capture the patient's actions when executing the exercise prescription through a monitoring device and compare them based on a standard action template; Evaluate the quality of the patient's actions and generate feedback information according to the deviation value; Dynamically adjust the exercise prescription according to the patient's training effect and deviation analysis results.

[0014] The present invention also provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the above-mentioned method is implemented.

[0015] The present invention also provides a storage medium, on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned method is implemented.

[0016] The present invention provides a cardiac rehabilitation system, method, device, and storage medium for patients with acute myocardial infarction. It has the following Beneficial effects: 1. The present invention combines the patient's static data (such as basic information, BMI, cardiopulmonary function, etc.) and dynamic data (such as real-time heart rate, movement speed, etc.), and uses an optimization algorithm to generate a highly personalized exercise prescription. This prescription is dynamically adjusted according to the patient's actual rehabilitation needs and health status, effectively avoiding the "one-size-fits-all" problem in traditional rehabilitation methods and improving the adaptability and scientific nature of the rehabilitation plan.

[0017] 2. The motion monitoring module of the present invention can accurately evaluate the quality of the patient's motion execution by capturing the patient's skeletal key points in real time and comparing them with the standard motion template. When the patient's motion deviates significantly from the standard template, the system will automatically generate corrective suggestions and push them to the patient side. This mechanism effectively avoids the decline in training effect or rehabilitation risk caused by incorrect motions and improves the correct rate of rehabilitation motions.

[0018] 3. The feedback adjustment module of the present invention dynamically adjusts the content of the exercise prescription (including exercise type, intensity, time, and frequency) according to the analysis results of the patient's training effect and motion deviation. This real-time feedback and adjustment mechanism ensures that the rehabilitation plan can be optimized according to the patient's real-time performance, enabling the rehabilitation training to always be carried out within a safe and effective range, thereby improving the sustainability of the rehabilitation effect.

[0019] 4. The present invention provides an intuitive and easy-to-use display of the training plan, motion feedback, and corrective guidance for patients through the user interaction module, facilitating the patients to better participate in the rehabilitation training. At the same time, it provides convenient patient file management and exercise prescription adjustment functions for medical staff, improving the work efficiency of medical staff and realizing the collaborative management between doctors and patients.

[0020] 5. The present invention forms a complete closed-loop rehabilitation management process through the modular design of data collection, exercise prescription generation, motion monitoring, feedback evaluation, and dynamic adjustment. The data between each module is interacted in real time, ensuring that the patient's rehabilitation plan is a complete closed loop from design to execution, which not only improves the rehabilitation efficiency but also reduces the rehabilitation risk caused by non-scientific training for patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 is a schematic diagram of the system architecture of the present invention; Figure 2 is a schematic diagram of the method flow of the present invention; Figure 3 is a schematic diagram of the structure of the computer device of the present invention.

[0022] Among them, 40, computer device; 41, processor; 42, memory; 43, storage medium. DETAILED DESCRIPTION OF THE INVENTION

[0023] Next, in combination with the accompanying drawings of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0024] Please refer to the attached Figure 1 , the present invention provides a cardiac rehabilitation system for patients with acute myocardial infarction, aiming to provide personalized and precise rehabilitation services for patients with acute myocardial infarction. Through intelligent technical means, it realizes the generation of exercise prescriptions, action monitoring and dynamic adjustment, and improves the rehabilitation effect of patients. The system of the present invention includes a data processing module, an exercise optimization module, an action monitoring module, a feedback adjustment module and a user interaction module. The following will detail each module of the system of the present invention.

[0025] Data processing module In this embodiment, the data processing module is used to collect, process and store the static and dynamic data of patients. The static data includes basic data such as the basic information of patients and exercise cardiopulmonary function parameters; the dynamic data includes real-time indicators during the rehabilitation training of patients, such as heart rate, blood pressure, exercise speed, energy consumption, etc. It should be noted that these data are important inputs for subsequent exercise prescription generation and training quality evaluation, so the integrity, accuracy and real-time nature of the data are crucial.

[0026] In a possible implementation manner, the data processing module works in cooperation through multiple data acquisition subsystems, including: a static data acquisition sub-module, a dynamic data acquisition sub-module and a data storage sub-module. These sub-modules communicate through a unified data interface, so as to achieve seamless transmission and processing of data.

[0027] Specifically, the static data acquisition sub-module is used to collect the basic health information and exercise ability evaluation data of patients. Exemplarily, the basic health information of patients includes age, gender, height, weight, body mass index (BMI), past medical history, etc. As an option, the body mass index can be calculated by the following formula: where W represents the patient's weight (in kilograms), and H 2 represents the patient's height (in meters).

[0028] It should be noted that BMI is an important parameter for modeling the health status of patients and can be used to evaluate the body composition status of patients and its impact on exercise tolerance.

[0029] In some embodiments, the static data further includes key exercise ability parameters obtained through a cardiopulmonary exercise test (CPET), such as peak oxygen uptake (VO 2 peak) and maximum heart rate (HR max ). Among them: VO 2 peak is used to reflect the cardiopulmonary function of the patient and is directly related to the patient's exercise tolerance; HR max is an important reference parameter for exercise intensity setting, and it can be estimated by the formula: HR max = 220 - A where A is the age of the patient.

[0030] It can be understood that the above static data can be directly entered through the hospital information system or supplemented and obtained by the patient filling it in themselves.

[0031] In this embodiment, the dynamic data acquisition sub-module is used to acquire the real-time data of the patient during training, including heart rate, blood pressure, exercise speed, energy consumption, electrocardiogram parameters, etc. These data are acquired by wearable devices (such as smart bracelets or electrocardiogram monitoring devices) and wirelessly transmitted to the data processing module.

[0032] In a possible implementation, the heart rate data is acquired at a frequency of once per second and is used to evaluate the patient's real-time exercise load. Exemplarily, the acquired heart rate data range needs to meet the following constraint conditions: HR(t) ∈ [0.6·HR max , 0.8·HR max where HR(t) is the real-time heart rate of the patient at a certain moment.

[0033] As an option, the energy consumption data can be estimated based on the exercise speed and body weight. Specifically, the energy consumption can be calculated by the formula: E(t) = k·W·v(t) where k is the energy conversion coefficient related to the exercise type, v(t) is the real-time exercise speed of the patient, and W is the patient's body weight.

[0034] It should be noted that the electrocardiogram parameter (ECG(t)) is used to monitor the real-time cardiac activity status of the patient and identify possible abnormalities during training, such as arrhythmia or myocardial ischemia.

[0035] In some embodiments, in order to improve the reliability of data acquisition, the wearable device is equipped with multi-modal sensors, such as accelerometers, heart rate sensors, and pressure sensors, to ensure the accuracy and integrity of the acquired data. ​

[0036] In this embodiment, the collected static and dynamic data are transmitted to the data storage sub-module through a unified interface. The data storage sub-module is managed by a central database and provides data support for the subsequent motion optimization module and action monitoring module.

[0037] As a possible implementation, the data storage sub-module stores data in a hierarchical manner: Static data (such as patient basic information and exercise ability assessment data) is stored in the user profile layer of the system; Dynamic data (such as real-time heart rate, energy consumption, etc.) is stored in the training data layer and indexed according to timestamps.

[0038] To ensure data security, an encrypted transmission protocol (such as TLS) and a data encryption algorithm (such as AES) are used in this embodiment to protect the patient's personal data.

[0039] It should be noted that the data storage sub-module has a data cleaning function and can automatically filter or complete outliers (such as device acquisition errors or transmission packet losses) in the collected data.

[0040] In this embodiment, the data processing module outputs the processed data to the motion optimization module and the action monitoring module through a standardized interface. As an option, the output data format uses JSON or XML to improve data compatibility between different modules.

[0041] Exemplarily, the output data content includes: Static data (such as the patient's age, BMI, peak oxygen uptake); Dynamic data (such as real-time heart rate, energy consumption, electrocardiogram parameters, etc.).

[0042] It can be understood that the efficient operation of the data processing module can provide high-quality data support for subsequent modules, thereby improving the operation effect of the entire system.

[0043] Motion Optimization Module In this embodiment, the motion optimization module is used to generate a personalized exercise prescription based on the patient's static data and dynamic data. The exercise prescription includes exercise type, exercise intensity, exercise time, and exercise frequency, and ensures the scientificity and adaptability of the exercise plan through an optimization algorithm. It should be noted that the motion optimization module is the core module of the entire rehabilitation system, and the exercise prescription it generates directly determines the content of the patient's rehabilitation plan.

[0044] In a possible implementation, the motion optimization module includes a health modeling sub-module, an optimization solving sub-module, and a constraint condition module. These sub-modules work together through a unified logical process to achieve the dynamic generation and optimization of the exercise prescription.

[0045] In this embodiment, the health modeling sub-module constructs a health target model based on the static data and dynamic data of the patient. The health target model is used to quantify the rehabilitation targets such as the cardiopulmonary function, muscle strength, and balance ability of the patient, and provide an optimization basis for the generation of exercise prescriptions.

[0046] Specifically, the health target model is defined as an objective functional, which is used to represent the trade-off relationship among multiple targets in the rehabilitation process. Exemplarily, the objective functional can be expressed as: Where: u(t) represents the exercise parameters for each time period in the exercise prescription, including exercise type, intensity, and duration; f 1 (u(t)) is the cardiopulmonary function improvement function, which is used to reflect the change in the patient's oxygen utilization ability; f 2 (u(t)) is the muscle strength improvement function, which is used to represent the improvement of the patient's muscle endurance by strength training; f 3 (u(t)) is the balance ability improvement function, which is used to evaluate the improvement of the patient's stability by balance training; α 1 ,α 2 ,α 3 are weight parameters, which reflect the importance of each rehabilitation target.

[0047] As an option, the cardiopulmonary function improvement function f 1 (u(t)) can be defined based on the patient's peak oxygen uptake (VO 2 peak), and the specific expression is: Where, ΔVO 2 peak is the change in oxygen consumption induced by training.

[0048] It should be noted that the muscle strength improvement function f 2 (u(t)) can be expressed in the form of a quadratic function: f 2 (u(t)) = k 1 u(t) 2 - k 2 u(t) Where, κ 1 and k 2 are personalized parameters, which respectively represent the incremental coefficient and resistance coefficient of strength improvement.

[0049] The balance ability improvement function f 3(u(t)) can be defined based on the patient's range of joint motion and motion coordination index.

[0050] It can be understood that the health modeling sub-module provides a mathematical description for the generation of exercise prescriptions through weighted combination of the above functions, and dynamically adjusts the weight parameters in combination with the patient's personalized needs.

[0051] In this embodiment, the optimization and solution sub-module generates an optimal exercise prescription based on the health target model through an optimization algorithm. Specifically, the optimization objective is to minimize the objective functional J(u), that is: u * (t) = argmin J(u) where u * (t) represents the optimal exercise parameters, including exercise type, intensity, and time.

[0052] As an implementation, the optimization and solution sub-module uses the variational method to solve the extremum of the objective functional. The variational method combines the objective function with the constraint conditions by introducing Lagrange multipliers to form an optimization model. The specific expression is: where: g i (u(t)) represents the constraint conditions, including restrictions on exercise intensity, time, and heart rate, etc.; λ i is the Lagrange multiplier, representing the weight of the constraint conditions.

[0053] It should be noted that the optimization and solution sub-module calculates the optimal solution by solving the Euler-Lagrange equation: Combined with the initial condition u(0) and the boundary condition u(T), the exercise parameters for each time period are calculated.

[0054] In some embodiments, the optimization and solution sub-module can also be adjusted in real time according to the patient's dynamic data to ensure the scientific nature and dynamic adaptability of the exercise prescription.

[0055] In this embodiment, the constraint condition module is used to limit the intensity, time, and physiological indicators of the exercise prescription to ensure the safety and applicability of the exercise.

[0056] Specifically, the constraint conditions include the following aspects: Heart rate limit: HR(t) ∈ [0.6·HR max , 0.8·HR max where HR max ​$HR_{max}$ is the maximum heart rate of the patient, and its calculation method is described in the aforementioned data processing module.

[0057] Exercise intensity limit: $u(t)\leq u$ max $u$ max is the upper limit of the safe exercise intensity for the patient, which is related to the patient's basal metabolic rate and energy consumption ability.

[0058] Exercise time constraint: where $T$ max represents the maximum duration of each exercise.

[0059] It can be understood that the setting of the constraint condition module is based on the patient's physiological state and rehabilitation needs, and can effectively prevent the occurrence of excessive or insufficient exercise.

[0060] In this embodiment, the output of the exercise optimization module is a personalized exercise prescription. Exemplarily, the content of the exercise prescription includes: Exercise type: such as aerobic exercise (fast walking, cycling, etc.), resistance training (weight training), flexibility training (stretching exercise), and balance training (standing balance training).

[0061] Exercise intensity: Dynamically adjusted based on the patient's heart rate and muscle endurance level.

[0062] Duration of each exercise: Usually 20 - 40 minutes, and the specific time is determined according to the optimization result.

[0063] Training frequency per week: At least 3 times, and recommended 5 times.

[0064] It should be noted that the exercise prescription is transmitted to the subsequent module through the data interface and serves as the input basis for the action monitoring module.

[0065] Action monitoring module In this embodiment, the action monitoring module is used to capture the patient's actions in real time when executing the exercise prescription, monitor the accuracy and standardization of the action execution, and generate corrective suggestions based on the comparison result between the action deviation and the standard template. It should be noted that the action monitoring module collaborates with multiple technical means to achieve high-precision monitoring and evaluation of the patient's rehabilitation training and ensure the correct implementation of the exercise prescription.

[0066] In a possible implementation manner, the action monitoring module includes an action capture sub-module, a standard comparison sub-module, and a deviation detection sub-module. Each sub-module collaborates with each other in a logical order to implement the complete process from action collection to the output of corrective suggestions.

[0067] In this embodiment, the motion capture sub-module is used to collect the skeletal motion data of the patient during the movement process and construct a skeletal key point model of the patient. Specifically, this module can be implemented by a camera device or a wearable motion capture device.

[0068] As an option, the camera device can use a high-frame-rate RGB camera and combine it with a key point detection algorithm (such as the Pose Estimation model in Pytorch) to extract the skeletal key point data of the patient. The skeletal key points include the main joint positions such as the shoulders, elbows, knees, and hips. Exemplarily, these key points form a human skeletal structure diagram G patient =(V, E), where: V represents the set of skeletal key points, including all joint positions; E represents the connection relationship between the skeletal key points (such as the line connecting the shoulder joint and the elbow joint).

[0069] It should be noted that the accuracy of the skeletal key point model is crucial for the subsequent motion evaluation results. Therefore, during the motion capture process, the resolution of the camera device, the lighting conditions, and the posture changes of the patient need to be optimized to reduce errors.

[0070] In some embodiments, in order to improve the accuracy of motion capture, wearable devices (such as IMU sensors, accelerometers) are used in combination with camera devices. This method improves the spatio-temporal consistency of the skeletal key point data through multi-sensor fusion technology.

[0071] In this embodiment, the standard comparison sub-module is used to compare the patient's skeletal motion model G patient with the standard action template G standard pre-stored in the system to evaluate the accuracy and standardization of the patient's actions.

[0072] Specifically, the standard action template G standard is generated by recording the standard actions of healthy people or professional trainers. When generating the template, the system uses the same skeletal key point detection method to extract the skeletal structure of the standard action and stores it as a graph model G standard =(V ′ , E ′ ).

[0073] In a possible implementation manner, the comparison between the patient's action and the standard template is performed through a graph convolutional network (GCN). The GCN can effectively process the graph structure data of the skeletal key points and extract the local and global features between the patient's action and the standard template. Exemplarily, after the patient's skeletal graph and the standard skeletal graph are input into the GCN model, an action similarity score S is output, and its value range is from 0 to 1. The higher the score, the closer the action is to the standard template.

[0074] It is understandable that the standard comparison sub-module assigns higher weights to specific key points during the comparison process. For example, in balance training, the key point errors of the hip and knee joints have a greater impact on the results, so these key points will be processed preferentially.

[0075] In this embodiment, the deviation detection sub-module is used to generate action correction suggestions based on the deviation between the patient's action and the standard action.

[0076] Specifically, the deviation detection is achieved by calculating the spatial deviation between the key points of the patient's action and the key points of the standard action. Exemplarily, the deviation metric formula can be expressed as: Where: represents the three-dimensional coordinates of the i-th key point in the patient's action; represents the three-dimensional coordinates of the corresponding key point in the standard template; N is the total number of key points.

[0077] It should be noted that when the deviation value D exceeds the set threshold δ, the system will determine that the patient's action does not meet the standard and trigger the correction mechanism.

[0078] In a possible implementation, the deviation detection result can also be further analyzed through a specific algorithm. For example, specific incorrect key points (such as insufficient knee bending angle) can be identified through local sensitivity detection, so as to output more accurate correction suggestions.

[0079] In this embodiment, the final output of the action monitoring module is the score of the action execution quality and the corresponding correction suggestions. As an implementation, the score and suggestions are displayed through the patient-side interface, and the specific content includes: Action execution score: Calculated based on the similarity score S and the deviation value D, ranging from 0 to 100 points; Action correction suggestions: Provide correction methods for errors in specific key points, such as "insufficient knee joint angle, need to increase the bending amplitude".

[0080] It is understandable that this output form facilitates the patient to timely understand the action quality and adjust the action according to the suggestions, thereby improving the rehabilitation effect.

[0081] It should be noted that the operation process of the action monitoring module includes the following steps: First, the action capture sub-module obtains the real-time action data of the patient through a camera or wearable device and generates a skeletal key point model; Next, the standard comparison sub-module inputs the patient's action model into the GCN, compares it with the standard action template, and generates an action similarity score; Then, the deviation detection sub-module calculates the spatial deviation between the patient's movement and the standard template, and identifies the key error points in the movement; finally, the movement monitoring module outputs the movement execution quality score and correction suggestions, and displays them to the patient through the user interaction module.

[0082] In this embodiment, the movement monitoring module combines the bone key point detection technology and the graph convolutional network to achieve high-precision monitoring and standardized evaluation of the patient's movement. By capturing the patient's real-time movement data, comparing it with the standard movement template, generating deviation results and correction suggestions, this module provides guarantee for the patient's rehabilitation training.

[0083] Feedback adjustment module In this embodiment, the feedback adjustment module is used to dynamically adjust the exercise prescription according to the patient's rehabilitation training effect and movement monitoring results, so as to ensure the scientificity and effectiveness of the rehabilitation plan. It should be noted that the feedback adjustment module generates new movement parameters by collecting and analyzing the deviation data and health status changes during the patient's training, and updates the exercise prescription in real time.

[0084] In a possible implementation manner, the feedback adjustment module includes a deviation analysis sub-module and a prescription adjustment sub-module. The two sub-modules work together to generate a dynamic rehabilitation plan adapted to the patient's current health status through comprehensive evaluation of the training deviation.

[0085] In this embodiment, the deviation analysis sub-module is used to calculate the deviation between the implementation effect of the current exercise prescription and the target health status based on the patient's training effect data and movement monitoring results.

[0086] Specifically, this sub-module first receives the deviation data from the movement monitoring module, including the patient's movement execution deviation value D and the execution quality score S.

[0087] It can be understood that the deviation value D can quantify the difference between the patient's movement and the standard movement, thus providing a specific basis for dynamic adjustment.

[0088] In some embodiments, the deviation analysis sub-module also combines the patient's dynamic health indicators (such as heart rate, blood pressure, energy consumption, etc.) to calculate the actual deviation of the training effect through the health objective function. Exemplarily, the health objective deviation can be expressed as: ΔJ=|J actual (u)-J target (u)| Where: J actual (u) is the actual health status value of the patient after training; J target (u) is the target health status value of the exercise prescription.

[0089] It should be noted that when the health target deviation ΔJ exceeds the preset threshold, the system will initiate a dynamic adjustment process to update the exercise parameters.

[0090] In this embodiment, the prescription adjustment sub-module is used to dynamically adjust the key parameters of the exercise prescription (including exercise type, intensity, time, and frequency) according to the deviation analysis result to optimize the rehabilitation effect of the patient.

[0091] In a possible implementation, the prescription adjustment sub-module iteratively optimizes the exercise parameters by the gradient descent method. Specifically, the new exercise parameter u ′ (t) is calculated by the following formula: Where: u(t) is the current exercise parameter; η is the learning rate, which controls the step size of parameter adjustment; is the partial derivative of the objective function with respect to the exercise parameter, reflecting the impact of the current training parameter on the health target.

[0092] As an option, the prescription adjustment sub-module will dynamically adjust the optimization direction according to the weights of different rehabilitation goals. For example, for patients who need to prioritize improving cardiopulmonary function, the system will increase the proportion of aerobic exercise and reduce the intensity of resistance training.

[0093] In some embodiments, the prescription adjustment sub-module will also generate a personalized adjustment plan by combining the patient's training performance and historical health data records. For example, if it is detected that the patient's exercise heart rate is consistently low, the system will appropriately increase the upper limit of the exercise intensity and update the heart rate limit range.

[0094] It can be understood that the prescription adjustment sub-module can not only perform dynamic optimization based on the current training effect, but also make predictive adjustments to the patient's future rehabilitation needs through trend analysis of historical data.

[0095] In this embodiment, the working process of the feedback adjustment module includes the following steps: First, the deviation analysis sub-module receives the deviation data from the motion monitoring module and the patient's dynamic health indicators, and calculates the health target deviation ΔJ; Then, the prescription adjustment sub-module adjusts the exercise parameters using the optimization algorithm according to the deviation analysis result to generate a new exercise prescription; Finally, the updated exercise prescription is pushed to the patient side and the medical staff side through the user interaction module for the patient to execute and for the medical staff to refer to.

[0096] In a possible implementation, the feedback adjustment module dynamically changes the adjustment frequency according to the different rehabilitation stages of the patient. For example, in the initial stage of rehabilitation, the adjustment frequency can be set to be updated daily; while in the later stage of rehabilitation, it can be adjusted to be updated once a week to avoid disturbing the patient due to frequent adjustments.

[0097] In this embodiment, the final output of the feedback adjustment module is the updated exercise prescription. Exemplarily, the content of the new exercise prescription includes: The adjusted exercise type, for example, increasing the proportion of balance training to improve the patient's stability; The updated exercise intensity, for example, the exercise intensity range optimized based on the health goal; The modified exercise time and frequency, for example, extending the single training time to 40 minutes and increasing the weekly training frequency to 5 times.

[0098] It can be understood that these adjustment results can further optimize the patient's rehabilitation effect in subsequent training, and at the same time provide new input data for other modules (such as the motion monitoring module).

[0099] It should be noted that the operation of the feedback adjustment module depends on the input data from the motion monitoring module and the data processing module, and transfers the adjusted exercise prescription as the output to the user interaction module and the motion optimization module, thus forming a closed-loop dynamic feedback mechanism.

[0100] In some embodiments, the feedback adjustment module also records the process and results of each adjustment for medical staff to review and analyze at the management end.

[0101] User Interaction Module In this embodiment, the user interaction module is used for the interaction between the patient and the system, supporting the patient to receive the exercise prescription, view the training feedback, obtain action correction suggestions, and at the same time providing functions such as patient file management, rehabilitation progress query, and exercise prescription adjustment for medical staff. It should be noted that the user interaction module is implemented through two sub-modules, the patient end and the medical staff end, which provide different functions for patients and medical staff respectively.

[0102] In a possible implementation, the user interaction module is implemented in a lightweight manner. The patient end uses a WeChat mini-program, and the medical staff end uses a Web management system based on the B / S architecture. This design method reduces the system usage threshold and improves the usability and efficiency of the module at the same time.

[0103] Patient End In this embodiment, the patient end is mainly for patients with acute myocardial infarction, providing functions such as display of rehabilitation plans, training feedback, and real-time guidance. The core functions of the patient end include display of exercise prescriptions, viewing of real-time feedback, push of action demonstrations and correction suggestions.

[0104] As an option, the patient terminal presents the specific content of the exercise prescription to the patient through a simple user interface, including: Exercise type: such as aerobic exercise, resistance training, flexibility training, and balance training; Exercise intensity: presenting the target heart rate range in text or chart form; Exercise time and frequency: for example, the exercise time per session is set to 30 minutes, and the training frequency per week is set to 5 times.

[0105] In some embodiments, the patient terminal also provides a real-time feedback function, displaying dynamic data such as the patient's training completion rate, exercise intensity, and energy consumption through charts. For example, the patient can view the comparison between the heart rate curve and the target heart rate interval in real time to understand whether the training intensity meets the expectations.

[0106] Specifically, the patient terminal also supports the function of action demonstration and correction. When the action monitoring module detects a deviation in the patient's action, the system will push correction suggestions through the patient terminal. For example, when it is detected that the knee bending angle is insufficient, a prompt will pop up on the patient terminal interface: "The knee bending is insufficient. Please bend the knee to less than 90 degrees." At the same time, the patient terminal will also provide a standard action demonstration video to guide the patient to perform the training actions correctly.

[0107] In a possible implementation, the patient terminal also supports the training plan reminder function. Through the message push mechanism of the mini-program, the patient terminal will push training reminders before the daily training time arrives. For example, the system will send a message: "Today is a rehabilitation training day. Please complete 30 minutes of brisk walking training according to the plan."

[0108] It should be noted that the data update and display on the patient terminal are real-time, and all training data are synchronously updated from other modules through the background interface.

[0109] Medical staff terminal In this embodiment, the medical staff terminal is mainly for rehabilitation doctors, exercise prescription doctors, and other medical staff, providing functions such as patient file management, training progress query, and exercise prescription adjustment. The core goal of the medical staff terminal is to provide a convenient and efficient patient management tool for medical staff.

[0110] Specifically, the medical staff terminal supports viewing the patient's rehabilitation data through a Web interface, including the patient's basic information, training records, and the changing trends of health indicators. For example, medical staff can view the patient's cardiopulmonary tolerance, muscle strength, and dynamic heart rate data, and analyze the patient's rehabilitation effect through a curve graph.

[0111] In a possible implementation, the medical staff side supports the patient file management function, and all static and dynamic data of patients can be quickly retrieved through the file module. Exemplarily, the content displayed by the file module includes: Basic patient information: such as name, age, gender, BMI, etc.; Summary of the rehabilitation plan: such as the content of the current exercise prescription, including exercise type, intensity, and time; Historical data: such as the patient's training completion rate, health goal deviation ΔJ, etc.

[0112] It can be understood that the medical staff side also supports the manual adjustment function of the exercise prescription. When the feedback adjustment module generates a new exercise prescription, medical staff can review and confirm the adjustment content through the medical staff side. As an option, the medical staff side supports a drag-and-drop adjustment interface, and medical staff can quickly modify the exercise type or adjust the intensity range through simple drag-and-drop operations.

[0113] In some embodiments, the medical staff side also supports the batch management function of the rehabilitation plan. For example, for multiple patients in the same disease course stage, the system allows medical staff to apply the same adjustment plan with one click, thereby improving the management efficiency.

[0114] It should be noted that all operations on the medical staff side are controlled by the permission management module to ensure the privacy and security of patient data and the traceability of operations.

[0115] In this embodiment, both the patient side and the medical staff side of the user interaction module support the data visualization function, and the patient's rehabilitation progress is displayed through charts and data cards.

[0116] As an option, the content of the data display includes: Training completion rate: for example, "The training completion rate this week is 85%, and 4 trainings have been completed"; Action execution score: Based on the scoring results of the action monitoring module, provide an action execution score of 0-100 points; Energy consumption trend: for example, "The total energy consumption this week is 1500 kcal, an increase of 10% compared to last week".

[0117] It can be understood that the data visualization function can help patients more intuitively understand the rehabilitation effect, and at the same time provide a scientific basis for medical staff to evaluate the patient's recovery progress.

[0118] It should be noted that the user interaction module and other modules perform data interaction through a standardized interface. Specifically: The user interaction module obtains the static data of patients from the data processing module; Obtain training feedback data and action correction suggestions from the action monitoring module; Obtain the updated exercise prescription from the feedback adjustment module.

[0119] In a possible implementation, the user interaction module communicates bidirectionally with the central database, and all data updates are synchronized to the database in real time and pushed to the patient side and the medical staff side through the interface.

[0120] In this embodiment, the user interaction module combines the different needs of the patient side and the medical staff side, and realizes the guidance of the patient's rehabilitation training and the management support for the medical staff through a lightweight design.

[0121] Through the collaborative work of the data processing module, the motion optimization module, the motion monitoring module, the feedback adjustment module and the user interaction module, the present invention realizes the intelligent management of the whole process of the patient's rehabilitation. The system can collect the static and dynamic health data of the patient, generate a personalized exercise prescription based on the optimization algorithm, and dynamically adjust the prescription content through motion monitoring and real-time feedback to ensure the scientificity and safety of the rehabilitation training. The interactive design of the patient side and the medical staff side realizes the accurate push and efficient management of the rehabilitation plan, forming a data-driven closed-loop rehabilitation process. The present invention improves the effect and compliance of the home rehabilitation of patients with acute myocardial infarction, and provides a scientific and accurate solution for the field of heart disease rehabilitation.

[0122] Please refer to the attached Figure 2 , the present invention also provides a cardiac rehabilitation method for patients with acute myocardial infarction, including the following steps: Step S1: Collect the static data and dynamic data of the patient; In this embodiment, first, the static data and dynamic data of the patient are collected to provide the basic input for the generation of the exercise prescription and the training monitoring. The static data includes the basic information of the patient, such as age, gender, height, weight, BMI and past medical history. In addition, it also includes the cardiopulmonary function parameters (such as peak oxygen uptake and maximum heart rate) obtained through the cardiopulmonary exercise test.

[0123] The dynamic data is collected in real time through wearable devices, mainly including the heart rate, blood pressure, exercise speed, energy consumption and electrocardiogram parameters of the patient during the training process. These data are wirelessly transmitted to the central database and are called for subsequent steps.

[0124] As a possible implementation, the wearable device combines multi-modal sensors (such as accelerometers, heart rate sensors, etc.) to improve the accuracy of data collection, thereby ensuring the integrity and reliability of the data.

[0125] Step S2: Establish a health goal model and generate a personalized exercise prescription; In this embodiment, a health goal model is established using the static and dynamic data of the patient. This model is used to quantify the rehabilitation goals of the patient's cardiopulmonary function, muscle strength, and balance ability. The personalized exercise prescription generated based on the model includes exercise type, intensity, time, and frequency.

[0126] The exercise prescription is dynamically generated through an optimization method and can adapt to the individual differences and rehabilitation needs of the patient. Specifically, the system uses optimization algorithms such as the variational method to solve for the optimal exercise parameters according to the patient's current health status to ensure that the exercise intensity is appropriate and the time is reasonable.

[0127] It should be noted that after the exercise prescription is generated, it will be transmitted to the patient side through the user interaction module, and the patient can clearly view the specific training plan content through the interface.

[0128] Step S3: Capture and compare the actions of the patient when performing the exercise prescription; In this embodiment, the actions of the patient when performing the exercise prescription are captured in real time through a monitoring device (such as a camera or a wearable device). The system uses key point detection technology to construct the patient's skeletal model and compares it with the standard action template built into the system.

[0129] In a possible implementation, the standard action template is generated from the recorded actions of healthy individuals or professional trainers and contains standard data such as joint angles and action amplitudes. The matching degree between the patient's skeletal model and the standard template is analyzed through a graph convolutional network (GCN) to identify the deviations in the actions.

[0130] It should be noted that this process can quickly detect specific problems in the patient's actions, such as insufficient movement of certain joints or excessive action amplitudes, thereby providing a basis for generating subsequent feedback.

[0131] Step S4: Evaluate the action quality and generate feedback information; In this embodiment, the system evaluates the action quality based on the deviation value between the patient's action and the standard template and generates targeted feedback information. The deviation value is calculated from the coordinate differences between the patient's skeletal model and the standard template.

[0132] The system generates real-time feedback information based on the deviation detection results and presents it to the patient through the patient-side interface. For example, when it is detected that the patient's action amplitude is insufficient, the system will prompt the patient to increase the angle of certain actions and provide a video demonstration of the standard action for reference.

[0133] As an option, the content of the feedback information can be graded according to the severity of the deviation, such as "minor adjustment suggestions" or "severe deviation warnings", to help the patient better understand the problems in the training.

[0134] Step S5: Dynamically adjust the exercise prescription according to the feedback In this embodiment, according to the patient's training effect and deviation analysis results, the system dynamically adjusts the exercise prescription. The adjustment content includes exercise type, intensity, time and frequency to ensure that the patient's rehabilitation plan matches the actual ability.

[0135] In a possible implementation, the system calculates the difference between the actual value and the target value of the health goal through the feedback adjustment module to determine whether to adjust the exercise parameters. For example, if it is found that the patient's heart rate has not reached the target range for a long time, the system will appropriately increase the upper limit of the exercise intensity.

[0136] The adjusted exercise prescription will be synchronized to the patient side and the medical staff side. The patient can view the updated content in real time, and the medical staff can also review or further modify the adjustment content on the medical staff side.

[0137] The cardiac rehabilitation method for acute myocardial infarction patients of the present invention realizes the scientificity and personalization of the patient's rehabilitation plan through an orderly process of data collection, exercise prescription generation, movement monitoring, feedback evaluation and dynamic adjustment.

[0138] Please refer to the appendix Figure 3 , the present invention also provides a computer device 40, including: a processor 41 and a memory 42. The memory 42 stores a computer program executable by the processor. When the computer program is executed by the processor, the above method is executed.

[0139] The present invention also provides a storage medium 43. A computer program is stored on the storage medium 43. When the computer program is run by the processor 41, the above method is executed.

[0140] Among them, the storage medium 43 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (abbreviated as SRAM), electrically erasable programmable read-only memory (abbreviated as EEPROM), erasable programmable read-only memory (abbreviated as EPROM), programmable read-only memory (abbreviated as PROM), read-only memory (abbreviated as ROM), magnetic memory, flash memory, magnetic disk or optical disc.

[0141] Although embodiments of the present invention have been shown and described, those of ordinary skill in the art will appreciate that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. Cardiac rehabilitation system for patients with acute myocardial infarction, characterized in that: include: Data processing module, used to collect and store static and dynamic data of patients, including basic information of patients, exercise cardiopulmonary function parameters, real-time heart rate, blood pressure and energy consumption; The exercise optimization module is used to generate personalized exercise prescriptions for patients based on their static and dynamic data, including exercise type, intensity, time and frequency; The action monitoring module is used to monitor the patient's action in executing the exercise prescription, evaluate the patient's action quality and provide correction suggestions based on the deviation between the patient's action and the standard action template; Feedback adjustment module, used to dynamically adjust the exercise prescription according to the patient's training effect and movement deviation; The user interaction module is used to provide patients with exercise prescriptions and feedback information, and for medical staff to query patient files and adjust rehabilitation plans.

2. The cardiac rehabilitation system for patients with acute myocardial infarction according to claim 1, characterized in that: The motion optimization module comprises: The health modeling module is used to establish a patient health target model based on the patient's cardiopulmonary function parameters, muscle strength parameters, and balance ability parameters; The optimization solution module is used to generate the optimal exercise prescription through optimization method based on the patient's health goal model; The constraint condition module is used to set constraints on exercise intensity, heart rate range, joint range of motion and cardiovascular load to ensure the safety of exercise prescription.

3. The cardiac rehabilitation system for patients with acute myocardial infarction according to claim 2, characterized in that: The exercise optimization module uses an optimization method based on the calculus of variations to generate an exercise prescription; wherein the health goal model is constructed based on the patient's cardiopulmonary function, muscle strength and balance ability, and specifically includes the following goals: Improve exercise endurance by optimizing patients' cardiopulmonary function parameters; Adjust the intensity of exercise according to the patient's muscle strength parameters and mobility; Improve joint mobility and balance stability through balance training.

4. The cardiac rehabilitation system for patients with acute myocardial infarction according to claim 1, characterized in that: The action monitoring module comprises: Motion capture module, used to capture the patient's motion trajectory in real time through a camera or wearable device and generate a skeleton key point model; A standard comparison module is used to match the patient's skeletal key point model with the standard motion template; The deviation detection module is used to calculate the deviation value between the patient's action and the standard action based on the deviation measurement method and trigger corrective suggestions.

5. The cardiac rehabilitation system for patients with acute myocardial infarction according to claim 4, characterized in that: The action monitoring module performs action quality assessment based on a graph convolutional network, including the following steps: Generate a patient's skeletal key point map, where the key points include the spatial coordinates of the shoulder, knee and joints; The patient's skeletal key point map is input into the graph convolutional network and compared with the standard action template; The correctness of the patient's movements is determined based on the deviation threshold and correction suggestions are generated.

6. The cardiac rehabilitation system for patients with acute myocardial infarction according to claim 1, characterized in that: The feedback adjustment module comprises: Deviation analysis module, used to analyze the difference between the patient's movement deviation and the target health parameter; The prescription adjustment module is used to dynamically adjust the exercise type, exercise intensity and exercise time according to the analysis results.

7. The cardiac rehabilitation system for patients with acute myocardial infarction according to claim 1, characterized in that: The user interaction module comprises: The patient side is used to show patients the rehabilitation plan and feedback information, including movement demonstration, training completion rate and correction suggestions; The management side is used by medical staff to query patient data, check rehabilitation progress and adjust exercise prescriptions.

8. A cardiac rehabilitation method for patients with acute myocardial infarction, based on the cardiac rehabilitation system for patients with acute myocardial infarction according to any one of claims 1 to 7, characterized in that: The following steps are involved: Collect static and dynamic data of patients, including basic information of patients, cardiopulmonary function parameters and real-time training indicators; Establish health goal models based on patient data and generate personalized exercise prescriptions through optimization methods; The patient's movements when executing exercise prescriptions are captured through monitoring equipment and compared with standard movement templates; Evaluate the patient's movement quality and generate feedback information based on the deviation value; Dynamically adjust the exercise prescription based on the patient's training effect and deviation analysis results.

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to claim 8 is implemented.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to claim 8 is implemented.

Images