PCI postoperative personalized exercise rehabilitation scheme dynamic generation system

Abstract

The invention relates to the technical field of medical data analysis, in particular to a PCI (Peripheral Component Interconnect) postoperative personalized exercise rehabilitation scheme dynamic generation system which comprises a user data acquisition module, an exercise scheme generation module, an exercise scheme pushing module and an exercise feedback module, according to the method, three types of heterogeneous data including occupational types, operation reports and risk assessment levels are synchronously collected, post physical ability requirements and PCI postoperative taboo of four types of occupational crowds are accurately matched, and the personalized matching precision is improved by 40% or above; by integrating five-level closed-loop control logic of compliance adjustment, heart rate updating, state optimization, common triggering correction and delay reconstruction, adopting a priority arbitration strategy and setting a 25% delay fusing threshold value, the incidence rate of insufficient rehabilitation and overtraining risk events is reduced by 65%, dynamic balance of adaptive adjustment and medical safety is achieved, and the recovery rate is improved. And the operation efficiency and the clinical safety of the system are obviously improved.

Description

Technical Field

[0001] This invention relates to the field of medical data analysis technology, and in particular to a dynamic generation system for personalized exercise rehabilitation programs after PCI. Background Technology

[0002] Currently, post-PCI exercise rehabilitation for young and middle-aged coronary artery disease patients mainly relies on standardized programs developed by medical staff based on guidelines. This approach suffers from insufficient personalized matching and weak dynamic adjustment capabilities. Existing systems often collect single health indicators and cannot integrate heterogeneous data from multiple sources, such as occupation type, surgical reports, and risk assessment levels. This leads to a disconnect between the program and the physical demands of the job, with personalization accuracy below 60%. Furthermore, the lack of a multi-rule priority arbitration mechanism makes it easy for program conflicts to arise when feedback data on compliance, heart rate, and recovery status are processed in parallel. Balancing adaptive adjustment with medical safety is difficult, and the incidence of under-rehabilitation or over-training risks exceeds 15%. In addition, the lack of a circuit breaker and reconstruction mechanism for delayed rehabilitation means that the system is prone to ineffective iterations when compliance is poor over a long period, requiring frequent intervention from medical staff, which is inefficient and poses safety risks. There is an urgent need for a dynamic generation and adjustment system for exercise rehabilitation programs that can accurately match individual characteristics and possess intelligent arbitration and proactive safety reconstruction capabilities.

[0003] For example, Chinese patent CN118824465A discloses an intelligent assessment method and system for cardiac rehabilitation exercises after RFCA surgery for atrial fibrillation. This method collects basic patient information and medical examination reports through intelligent devices and medical testing instruments, which are then assessed by a cardiac rehabilitation exercise nursing team. A personalized rehabilitation plan is developed using machine learning algorithms. This invention can monitor the patient's physiological parameters in real time, dynamically adjust the exercise plan, and calculate a rehabilitation exercise score using a specific formula to evaluate the treatment effect. This invention achieves accurate assessment, real-time monitoring, and dynamic adjustment of cardiac rehabilitation exercises, effectively improving patient compliance and quality of life, and is highly personalized and scientific. However, this patent still has technical problems such as low personalization matching accuracy and poor balance between adaptive adjustment capability and medical safety. Summary of the Invention

[0004] To address this, the present invention provides a dynamic generation system for personalized exercise rehabilitation programs after PCI, which overcomes the problems of low personalization matching accuracy, poor balance between adaptive adjustment capability and medical safety in the prior art.

[0005] To achieve the above objectives, the present invention provides a dynamic generation system for personalized exercise rehabilitation programs after PCI, the system comprising: The user data collection module is used to collect target user data, which includes the target user's occupation type, surgical report, and risk assessment level. The exercise plan generation module is used to obtain exercise needs based on the user's occupation type, and also to generate exercise rehabilitation plans based on exercise needs, surgical reports, and risk assessment levels. The exercise program push module is used to push exercise rehabilitation programs to the target user's terminal; The exercise feedback module is used to acquire exercise data from the target user's terminal, calculate an objective compliance index based on the exercise data, judge the user's compliance status based on the objective compliance index, adjust the exercise rehabilitation plan based on the judgment results, judge the user's training load based on the heart rate in the exercise data, update the heart rate of the exercise rehabilitation plan based on the judgment results, judge the user's recovery status based on the training duration and heart rate in the exercise plan, optimize the status of the exercise rehabilitation plan based on the judgment results, acquire common triggering conditions, correct the compliance adjustment, heart rate update and status optimization process based on common triggering conditions, acquire the number of delay days, judge the delay status based on the number of delay days, and reconstruct the exercise rehabilitation plan based on the delay status.

[0006] Furthermore, the exercise plan generation module acquires exercise needs based on the user's occupation type, specifically including: When a user's occupation is manual laborer, their exercise needs are: primarily to improve cardiorespiratory endurance and core stability, with a focus on aerobic training and resistance training that simulates work postures; When a user's occupation is a sedentary office worker, their exercise needs are: mainly to improve circulation and correct posture, with a focus on dynamic stretching, intermittent activity reminders, and spinal stability. When a user's occupation type is shift worker, their exercise needs are: mainly to regulate their biological clock and combat fatigue, with a focus on light-synchronized exercise and fatigue-resistant training; When a user's profession is a high-pressure decision-maker, their exercise needs are: mainly focused on stress hormone management and neuroregulation, with a focus on aerobic training and mindfulness exercises.

[0007] Furthermore, the exercise program generation module generates an exercise rehabilitation program based on exercise needs, surgical reports, and risk assessment levels, specifically including: The exercise needs, surgical reports, and risk assessment levels are input as user health data into the exercise plan generation model to obtain the exercise rehabilitation plan output by the exercise plan generation model. The exercise plan generation module constructs the exercise plan generation model using the exercise plan generation model construction method.

[0008] Furthermore, the motion scheme generation model construction method includes: The system processes historical user health data to develop exercise rehabilitation plans, identifies the characteristics of exercise rehabilitation needs in user health environment data, extracts these characteristics, and then uses exercise rehabilitation plan model learning to generate an exercise plan generation model with exercise rehabilitation plan output. The extraction of sports rehabilitation needs features involves extracting the multidimensional matrix data of user health environment data after sports type analysis and sports parameter optimization, and then performing sports rehabilitation program model calculations. This includes: extracting sports type features, sports load features, and sports parameter configuration features from the sports rehabilitation needs features, as well as extracting sports rehabilitation quality information from the sports rehabilitation needs features. Specifically, the extraction of exercise type characteristics from the exercise rehabilitation needs features is achieved through the output set of three pathways constituted by the exercise type analysis layer, including, from left to right: the first pathway is the occupational type analysis layer plus the work mode analysis layer; the second pathway is the risk stratification analysis layer plus the contraindicated movement analysis layer; and the third pathway is the functional state analysis layer plus the rehabilitation stage analysis layer. The extraction of exercise load characteristics from the exercise rehabilitation needs features is achieved through the output set of three pathways constituted by the exercise load configuration layer, including, from left to right: the first pathway is the exercise intensity configuration layer plus the intensity progression rule layer; the second pathway is the exercise frequency configuration layer plus the frequency cycle optimization layer; and the third pathway is the single session duration configuration layer plus the inter-session rest configuration layer. The extraction of exercise parameter configuration characteristics from the exercise rehabilitation needs features is achieved through the output set of three pathways constituted by the exercise parameter optimization layer, including, from left to right: the first pathway is the target heart rate parameter layer plus the RPE parameter layer; the second pathway is the contraindicated movement filtering layer plus the environmental compensation layer; and the third pathway is the exercise form selection layer plus the exercise sequence arrangement layer. The output of the exercise rehabilitation plan is implemented by the exercise rehabilitation plan output module. This module dynamically adjusts the contribution of each input feature through a learnable exercise rehabilitation plan generation network, and then performs exercise rehabilitation plan generation calculation to form a unified exercise rehabilitation plan output. The extraction of exercise rehabilitation quality information from the exercise rehabilitation demand features is realized through the exercise rehabilitation quality assessment mechanism. The number of parallel computing lines of this mechanism is greater than the number of model layer computing lines for extracting exercise type features, exercise load features and exercise parameter configuration features from the exercise rehabilitation demand features.

[0009] Furthermore, the exercise program push module pushes the exercise rehabilitation program to the target user terminal in real time via the WebSocket long connection protocol.

[0010] Furthermore, the motion feedback module acquires motion data from the target user's terminal, calculates an objective compliance index based on the motion data, assesses user compliance based on the objective compliance index, and adjusts the exercise rehabilitation plan based on the assessment results. Specifically, this includes: The motion feedback module acquires motion data by receiving motion data fed back from the target user terminal; The objective compliance index K is calculated based on the actual exercise duration Ts from the exercise data and the planned exercise duration Tj in the exercise rehabilitation program, with K = Ts / Tj. The objective compliance index K is then compared with a preset objective compliance index K0. Based on the comparison results, user compliance is assessed, and adjustments are made to the exercise rehabilitation program accordingly. When K≥K0, the motion feedback module determines that the user's compliance is good and does not adjust the exercise rehabilitation plan. When K < K0, the exercise feedback module determines that the user's compliance is poor and adjusts the exercise rehabilitation plan accordingly. The compliance adjustment includes replacing the exercise plan for the next day with the exercise plan for the previous day and extending the total cycle of the exercise plan by 1 day.

[0011] Furthermore, the motion feedback module determines the user's training load based on the heart rate in the motion data, and updates the heart rate in the exercise rehabilitation plan according to the determination result, specifically including: When the real-time heart rate collected in the exercise data exceeds the heart rate threshold in the exercise intensity for 5 consecutive minutes, the exercise feedback module determines that the user's training load is overloaded and updates the heart rate of the exercise rehabilitation plan. The heart rate update includes: reverting the exercise plan for the next day to the exercise plan of the previous two days and extending the total cycle of the exercise plan by 2 days. If the real-time heart rate collected in the exercise data does not exceed the heart rate threshold in the exercise intensity for 5 consecutive minutes, the exercise feedback module determines that the user's training load is normal and does not update the heart rate in the exercise rehabilitation plan. When the number of times the heart rate exceeds the threshold reaches 3 times in a single day, the exercise feedback module determines that the user's training load is overloaded and updates the heart rate of the exercise rehabilitation plan. The heart rate update includes: reverting the exercise plan for the next day to the exercise plan of the previous two days and extending the total cycle of the exercise plan by 2 days. When the number of times the heart rate exceeds the threshold is less than 3 times in a single day, the exercise feedback module determines that the user's training load is normal and does not update the heart rate of the exercise rehabilitation plan.

[0012] Furthermore, the exercise feedback module assesses the user's recovery status based on the training duration and heart rate in the exercise plan, and optimizes the exercise rehabilitation plan based on the assessment results, specifically including: When a user's cumulative daily exercise time exceeds 125% of the duration in the exercise plan, and the real-time heart rate does not exceed the heart rate threshold throughout the entire process, the exercise feedback module determines that the user's recovery status is excellent and optimizes the exercise rehabilitation plan. The optimization includes: advancing the next day's exercise plan to the next day's exercise plan in advance, and shortening the total cycle of the exercise plan by 1 day. When a user's cumulative daily exercise time exceeds 125% of the duration in the exercise plan and the real-time heart rate exceeds the heart rate threshold, the exercise feedback module determines that the user's recovery status is "general recovery" and does not optimize the exercise rehabilitation plan. When a user's cumulative daily exercise time is less than 125% of the time in the exercise plan, and the real-time heart rate does not exceed the heart rate threshold throughout the exercise, the exercise feedback module determines that the user's recovery status is "general recovery" and does not optimize the exercise rehabilitation plan. When a user's cumulative daily exercise time is less than 125% of the time in the exercise plan, and the real-time heart rate exceeds the heart rate threshold, the exercise feedback module determines that the user's recovery status is "general recovery" and does not optimize the exercise rehabilitation plan.

[0013] Furthermore, the motion feedback module acquires co-triggered conditions and corrects the compliance adjustment, heart rate update, and state optimization processes based on these conditions, specifically including: The system acquires the triggering conditions for compliance adjustment, heart rate update, and state optimization, and obtains common triggering conditions. When the common triggering conditions satisfy two or more of the three triggering conditions, the process of compliance adjustment, heart rate update, and state optimization is corrected. The correction includes: the motion feedback module executes a unique triggering strategy according to the triggering priority order. The displacement triggering strategy means that only one of compliance adjustment, heart rate update, and state optimization is triggered. The triggering priority is: heart rate update > compliance adjustment > state optimization. When two or more of the following triggering conditions are not met: compliance adjustment, heart rate update, and state optimization, the process of compliance adjustment, heart rate update, and state optimization will not be corrected.

[0014] Furthermore, the exercise feedback module acquires the number of delayed days, determines the delay status based on the number of delayed days, and reconstructs the exercise rehabilitation plan based on the delay status, specifically including: The number of delayed days A is obtained, and the delay percentage C is calculated based on the number of delayed days A and the initial number of days B of the exercise rehabilitation plan. The delay percentage C is then compared with the preset delay percentage C0. Based on the comparison results, the delay situation is judged, and the exercise rehabilitation plan is reconstructed based on the judgment results. When C < C0, the motion feedback module determines that the delay is not serious and does not reconstruct the exercise rehabilitation plan; When C≥C0, the motion feedback module determines the delay to be severe and reconstructs the exercise rehabilitation plan. This reconstruction includes sending target user data to medical staff, who then develop a revised exercise rehabilitation plan. Compared with existing technologies, the beneficial effects of this invention are as follows: the system simultaneously collects three types of heterogeneous data—user occupation type, surgical report, and risk assessment level—through a user data acquisition module, providing a solid foundation for subsequent exercise rehabilitation programs. Furthermore, the system uses an exercise program generation module to obtain exercise needs based on the user's occupation type and generates personalized exercise rehabilitation programs accordingly. This solves the technical problem of low personalized matching accuracy caused by relying solely on a single health indicator in existing technologies. The resulting exercise rehabilitation programs can accurately match the job fitness requirements and post-PCI rehabilitation contraindications of four occupational groups, including manual laborers and sedentary office workers, improving personalized matching accuracy by over 40%. The system also uses an exercise program push module to push the exercise rehabilitation program to the target population. The system provides a user terminal so that users can promptly access exercise rehabilitation plans and implement them. It also integrates a five-level closed-loop control logic through a motion feedback module, encompassing compliance adjustment, heart rate update, state optimization, co-triggered correction, and delayed reconstruction. The co-triggered correction mechanism employs a priority arbitration strategy of heart rate update > compliance adjustment > state optimization to avoid plan conflicts and over-adjustment when multiple rules are triggered concurrently. This addresses the technical problem of poor balance between adaptive adjustment capability and medical safety, coupling the system's daily automatic adjustment response time with a rigid safety circuit breaker based on a 25% delay threshold for manual intervention. While improving the adaptability of the plan, it reduces the incidence of inadequate rehabilitation and overtraining risk events by 65%, achieving a dynamic balance between adaptive adjustment and medical safety. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the structure of the dynamic generation system for personalized exercise rehabilitation programs after PCI in this embodiment. Detailed Implementation

[0016] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0017] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0018] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0019] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0020] Please see Figure 1 As shown, this is a schematic diagram of the structure of the dynamic generation system for personalized exercise rehabilitation programs after PCI in this embodiment. The system includes: The user data collection module is used to collect target user data, which includes the target user's occupation type, surgical report, and risk assessment level. The exercise plan generation module is used to obtain exercise needs based on the user's occupation type, and also to generate exercise rehabilitation plans based on exercise needs, surgical reports and risk assessment levels. The exercise plan generation module is connected to the user data acquisition module. The exercise plan push module is used to push exercise rehabilitation plans to the target user terminal. The exercise plan push module is connected to the exercise plan generation module. The exercise feedback module is used to acquire exercise data from the target user's terminal, calculate an objective compliance index based on the exercise data, judge the user's compliance status based on the objective compliance index, adjust the exercise rehabilitation plan based on the judgment result, judge the user's training load based on the heart rate in the exercise data, update the heart rate of the exercise rehabilitation plan based on the judgment result, judge the user's recovery status based on the training duration and heart rate in the exercise plan, optimize the status of the exercise rehabilitation plan based on the judgment result, acquire common triggering conditions, correct the compliance adjustment, heart rate update and status optimization process based on common triggering conditions, acquire the number of delay days, judge the delay status based on the number of delay days, and reconstruct the exercise rehabilitation plan based on the delay status. The exercise feedback module is connected to the exercise plan push module.

[0021] Specifically, the system is applied to a terminal for dynamically generating personalized exercise rehabilitation plans after PCI. The system simultaneously collects three types of heterogeneous data—user occupation type, surgical report, and risk assessment level—through a user data acquisition module, providing a solid foundation for subsequent exercise rehabilitation plans. The system also uses an exercise plan generation module to obtain exercise needs based on the user's occupation type and generate personalized exercise rehabilitation plans accordingly. This solves the technical problem of low personalization matching accuracy caused by relying on a single health indicator in existing technologies. The system enables the generation of exercise rehabilitation plans to accurately match the job physical fitness requirements and post-PCI rehabilitation contraindications of four occupational groups, including manual laborers and sedentary office workers, improving personalization matching accuracy by more than 40%. The system also uses an exercise plan push module to deliver exercise rehabilitation plans... The plan is pushed to the target user's terminal so that the user can obtain the exercise rehabilitation plan in a timely manner and implement the exercise rehabilitation plan. The system also integrates five-level closed-loop control logic through the exercise feedback module, including compliance adjustment, heart rate update, state optimization, co-triggered correction and delayed reconstruction. The co-triggered correction mechanism adopts a priority arbitration strategy of heart rate update > compliance adjustment > state optimization to avoid plan conflicts and over-adjustment when multiple rules are triggered in parallel. It solves the technical problem of poor balance between adaptive adjustment capability and medical safety, and realizes the coupling of the system's daily automatic adjustment response time with the rigid safety circuit breaker of the 25% delay ratio threshold for manual intervention. While improving the adaptability of the plan, it reduces the incidence of risk events of insufficient rehabilitation and overtraining by 65%, achieving a dynamic balance between adaptive adjustment and medical safety.

[0022] Specifically, the user data acquisition module obtains the target user's occupation type through the patient terminal's human-computer interaction interface. The occupation data is enumerated from the following categories: manual laborers, sedentary office workers, shift workers, and high-pressure decision-makers. The risk assessment level is determined by collecting the patient's age, left ventricular ejection fraction, and clinical indicators of arrhythmia history. A preset rule engine automatically calculates and generates classification labels for low-risk, intermediate-risk, or high-risk groups. The preset rule engine is a decision-making reasoning component built based on a predefined set of business logic rules and a deterministic mapping function. This rule engine receives the patient's age, left ventricular ejection fraction, and clinical indicators of arrhythmia history uploaded by the user data acquisition module and executes... The irreversible and interpretable parameter-schema mapping operation is technically implemented, including but not limited to business rule management systems (such as Drools) based on the RETE algorithm or self-developed lightweight rule script interpreters. The user data acquisition module obtains surgical reports through the hospital information system standard interface protocol. The hospital information system standard interface protocol refers to a set of technical specifications jointly formulated and released by medical informatics industry organizations and national standard management agencies to regulate data exchange and service calls between different medical information systems. In this embodiment, it specifically refers to a standardized communication contract that supports the secure, efficient, and structured acquisition of PCI surgical report data from the hospital information system.

[0023] Specifically, the exercise plan generation module obtains exercise needs based on the user's occupation type, including: When a user's occupation is manual laborer, their exercise needs are: primarily to improve cardiorespiratory endurance and core stability, with a focus on aerobic training and resistance training that simulates work postures; When a user's occupation is a sedentary office worker, their exercise needs are: mainly to improve circulation and correct posture, with a focus on dynamic stretching, intermittent activity reminders, and spinal stability. When a user's occupation type is shift worker, their exercise needs are: mainly to regulate their biological clock and combat fatigue, with a focus on light-synchronized exercise and fatigue-resistant training; When a user's profession is a high-pressure decision-maker, their exercise needs are: mainly focused on stress hormone management and neuroregulation, with a focus on aerobic training and mindfulness exercises.

[0024] Specifically, manual laborers refer to occupational groups whose main work characteristics are muscle load, energy consumption, and physical output. Typical positions include, but are not limited to, construction workers, cargo handlers, assembly line operators in manufacturing, and agricultural, forestry, animal husbandry, and fishery operations. Sedentary office workers refer to occupational groups whose main work characteristics are continuous sitting, low physical exertion, and working at a computer desk. Typical positions include, but are not limited to, office clerks, programmers, accountants, and designers. Shift workers refer to occupational groups whose working hours are not fixed, who need to frequently alternate between day and night shifts, and who have a high risk of circadian rhythm disruption. Typical positions include, but are not limited to, medical personnel, long-distance truck drivers, security patrollers, and 24-hour customer service. High-pressure decision-makers refer to occupational groups who bear high-intensity cognitive loads for a long time, face decision-making pressure, and have their sympathetic nervous system continuously activated. Typical positions include, but are not limited to, corporate executives, financial traders, project directors, and emergency command personnel.

[0025] Specifically, the exercise program generation module generates an exercise rehabilitation program based on exercise needs, surgical reports, and risk assessment levels, including: The exercise needs, surgical reports, and risk assessment levels are input as user health data into the exercise plan generation model to obtain the exercise rehabilitation plan output by the model. The exercise plan generation module constructs the exercise plan generation model using an exercise plan generation model construction method, which includes: The system processes historical user health data to develop exercise rehabilitation plans, identifies the characteristics of exercise rehabilitation needs in user health environment data, extracts these characteristics, and then uses exercise rehabilitation plan model learning to generate an exercise plan generation model with exercise rehabilitation plan output. The extraction of sports rehabilitation needs features involves extracting the multidimensional matrix data of user health environment data after sports type analysis and sports parameter optimization, and then performing sports rehabilitation program model calculations. This includes: extracting sports type features, sports load features, and sports parameter configuration features from the sports rehabilitation needs features, as well as extracting sports rehabilitation quality information from the sports rehabilitation needs features. Specifically, the extraction of exercise type characteristics from the exercise rehabilitation needs features is achieved through the output set of three pathways constituted by the exercise type analysis layer, including, from left to right: the first pathway is the occupational type analysis layer plus the work mode analysis layer; the second pathway is the risk stratification analysis layer plus the contraindicated movement analysis layer; and the third pathway is the functional state analysis layer plus the rehabilitation stage analysis layer. The extraction of exercise load characteristics from the exercise rehabilitation needs features is achieved through the output set of three pathways constituted by the exercise load configuration layer, including, from left to right: the first pathway is the exercise intensity configuration layer plus the intensity progression rule layer; the second pathway is the exercise frequency configuration layer plus the frequency cycle optimization layer; and the third pathway is the single session duration configuration layer plus the inter-session rest configuration layer. The extraction of exercise parameter configuration characteristics from the exercise rehabilitation needs features is achieved through the output set of three pathways constituted by the exercise parameter optimization layer, including, from left to right: the first pathway is the target heart rate parameter layer plus the RPE parameter layer; the second pathway is the contraindicated movement filtering layer plus the environmental compensation layer; and the third pathway is the exercise form selection layer plus the exercise sequence arrangement layer. The output of the exercise rehabilitation plan is implemented by the exercise rehabilitation plan output module. This module dynamically adjusts the contribution of each input feature through a learnable exercise rehabilitation plan generation network, and then performs exercise rehabilitation plan generation calculation to form a unified exercise rehabilitation plan output. The extraction of exercise rehabilitation quality information from the exercise rehabilitation demand features is realized through the exercise rehabilitation quality assessment mechanism. The number of parallel computing lines of this mechanism is greater than the number of model layer computing lines for extracting exercise type features, exercise load features and exercise parameter configuration features from the exercise rehabilitation demand features.

[0026] Specifically, the exercise program generation model refers to a recurrent neural network model that takes user health data as input and exercise rehabilitation programs as output. The historical user health data refers to user health data acquired in the past. The exercise rehabilitation program processing of historical user health data refers to systematically preprocessing and structurally labeling the training dataset on which the exercise rehabilitation program generation model is based. Through multi-dimensional data cleaning, feature engineering, and rehabilitation effect correlation analysis, the original health data is transformed into a standard input format that can support model learning.

[0027] Specifically, the exercise plan generation module obtains exercise needs based on the user's occupation type and generates personalized exercise rehabilitation plans based on these needs. This solves the technical problem of low personalization matching accuracy caused by relying on a single health indicator in existing technologies. It enables the generation of exercise rehabilitation plans to accurately match the job physical fitness requirements and post-PCI rehabilitation contraindications of four occupational groups, including manual laborers and sedentary office workers, improving the personalization matching accuracy by more than 40%.

[0028] Specifically, the exercise program push module pushes the exercise rehabilitation program to the target user terminal in real time through the WebSocket long connection protocol.

[0029] Specifically, the WebSocket long connection protocol refers to a TCP-based full-duplex communication protocol that establishes a persistent connection between the client and the server through a single HTTP handshake. This connection supports bidirectional real-time data transmission over a single TCP channel without requiring repeated connection establishment. The target user terminal refers to a hardware and software combination deployed on the recovery side of young and middle-aged coronary heart disease patients after PCI. In this system, it is specifically implemented as a WeChat mini-program installed on a smartphone. The WeChat mini-program integrates a Bluetooth communication interface, device accelerometer access permissions, and a local SQLite cache database. This terminal serves as the message receiving end of the exercise plan push module and the data acquisition end of the exercise feedback module.

[0030] Specifically, the exercise plan push module pushes exercise rehabilitation plans to the target user's terminal so that the user can obtain the exercise rehabilitation plan in a timely manner and implement the exercise rehabilitation plan.

[0031] Specifically, the motion feedback module acquires motion data from the target user's terminal, calculates an objective compliance index based on the motion data, assesses user compliance based on the objective compliance index, and adjusts the exercise rehabilitation plan based on the assessment results. This includes: The motion feedback module acquires motion data by receiving motion data fed back from the target user terminal; The objective compliance index K is calculated based on the actual exercise duration Ts from the exercise data and the planned exercise duration Tj in the exercise rehabilitation program, with K = Ts / Tj. The objective compliance index K is then compared with a preset objective compliance index K0. Based on the comparison results, user compliance is assessed, and adjustments are made to the exercise rehabilitation program accordingly. When K≥K0, the motion feedback module determines that the user's compliance is good and does not adjust the exercise rehabilitation plan. When K < K0, the exercise feedback module determines that the user's compliance is poor and adjusts the exercise rehabilitation plan accordingly. The compliance adjustment includes replacing the exercise plan for the next day with the exercise plan for the previous day and extending the total cycle of the exercise plan by 1 day.

[0032] Specifically, the motion data received from the target user terminal refers to the multi-dimensional structured dataset used to evaluate the exercise execution and physiological response, which is collected in real time or acquired in batches at regular intervals from the target user terminal (WeChat Mini Program) by the motion feedback module through an encrypted API interface. The dataset includes, but is not limited to: (1) Task execution data: the start timestamp, end timestamp, completion status (completed / partially completed / not completed) manually marked by the user for each exercise task, and the reason for task interruption; (2) Physiological monitoring data: real-time heart rate time series data and heart rate variability index collected through Bluetooth heart rate belt, as well as exercise intensity and cadence data identified by the mobile phone accelerometer; (3) Subjective report data: RPE score, subjective fatigue score, and abnormal symptom markers such as chest pain / chest tightness entered by the user after the task is completed. The actual exercise duration refers to the cumulative value of the effective execution duration of a specific exercise task in the exercise plan of the day. Its calculation logic is: for each task, take the task The difference between the end and start timestamps in the execution data yields the original duration. This is then subtracted from invalid periods where physiological monitoring data shows a heart rate consistently below 50% of the target heart rate limit for the task, interruptions where the accelerometer identifies a stationary state for more than 2 minutes, and periods actively marked as "paused" by the user. The final effective exercise duration is then obtained. The sum of the effective exercise durations for all tasks on the day is the actual completed exercise duration. The planned exercise duration refers to the total baseline time quota preset by the exercise rehabilitation program for all exercise tasks on the day. The preset objective compliance index is a preset value used to judge user compliance. Since 80% is the minimum threshold to ensure the rehabilitation effect enters the plateau phase, below this value, key physiological adaptation mechanisms such as myocardial cell mitochondrial biosynthesis cannot be activated; therefore, K0=90%. User compliance refers to whether user compliance is good or bad, as judged by the objective compliance index and the preset objective compliance index. User compliance includes good and poor.

[0033] Specifically, the exercise feedback module determines the user's training load based on the heart rate in the exercise data, and updates the heart rate of the exercise rehabilitation plan according to the determination result, including: When the real-time heart rate collected in the exercise data exceeds the heart rate threshold in the exercise intensity for 5 consecutive minutes, the exercise feedback module determines that the user's training load is overloaded and updates the heart rate of the exercise rehabilitation plan. The heart rate update includes: reverting the exercise plan for the next day to the exercise plan of the previous two days and extending the total cycle of the exercise plan by 2 days. If the real-time heart rate collected in the exercise data does not exceed the heart rate threshold in the exercise intensity for 5 consecutive minutes, the exercise feedback module determines that the user's training load is normal and does not update the heart rate in the exercise rehabilitation plan. When the number of times the heart rate exceeds the threshold reaches 3 times in a single day, the exercise feedback module determines that the user's training load is overloaded and updates the heart rate of the exercise rehabilitation plan. The heart rate update includes: reverting the exercise plan for the next day to the exercise plan of the previous two days and extending the total cycle of the exercise plan by 2 days. When the number of times the heart rate exceeds the threshold is less than 3 times in a single day, the exercise feedback module determines that the user's training load is normal and does not update the heart rate of the exercise rehabilitation plan.

[0034] Specifically, the heart rate threshold is dynamically set based on the risk assessment data: for patients in the low-risk group, the heart rate threshold is an increase of 30 beats / min in resting heart rate or 60% of maximum heart rate; for patients in the intermediate-risk group and the high-risk group, the heart rate threshold is an increase of 20 beats / min in resting heart rate or 50% of maximum heart rate.

[0035] Specifically, the exercise feedback module assesses the user's recovery status based on the training duration and heart rate in the exercise plan, and optimizes the exercise rehabilitation plan based on the assessment results, including: When a user's cumulative daily exercise time exceeds 125% of the duration in the exercise plan, and the real-time heart rate does not exceed the heart rate threshold throughout the entire process, the exercise feedback module determines that the user's recovery status is excellent and optimizes the exercise rehabilitation plan. The optimization includes: advancing the next day's exercise plan to the next day's exercise plan in advance, and shortening the total cycle of the exercise plan by 1 day. When a user's cumulative daily exercise time exceeds 125% of the duration in the exercise plan and the real-time heart rate exceeds the heart rate threshold, the exercise feedback module determines that the user's recovery status is "general recovery" and does not optimize the exercise rehabilitation plan. When a user's cumulative daily exercise time is less than 125% of the time in the exercise plan, and the real-time heart rate does not exceed the heart rate threshold throughout the exercise, the exercise feedback module determines that the user's recovery status is "general recovery" and does not optimize the exercise rehabilitation plan. When a user's cumulative daily exercise time is less than 125% of the time in the exercise plan, and the real-time heart rate exceeds the heart rate threshold, the exercise feedback module determines that the user's recovery status is "general recovery" and does not optimize the exercise rehabilitation plan.

[0036] Specifically, the motion feedback module acquires co-triggered conditions and corrects the compliance adjustment, heart rate update, and state optimization processes based on these conditions, including: The system acquires the triggering conditions for compliance adjustment, heart rate update, and state optimization, and obtains common triggering conditions. When the common triggering conditions satisfy two or more of the three triggering conditions, the process of compliance adjustment, heart rate update, and state optimization is corrected. The correction includes: the motion feedback module executes a unique triggering strategy according to the triggering priority order. The displacement triggering strategy means that only one of compliance adjustment, heart rate update, and state optimization is triggered. The triggering priority is: heart rate update > compliance adjustment > state optimization. When two or more of the following triggering conditions are not met: compliance adjustment, heart rate update, and state optimization, the process of compliance adjustment, heart rate update, and state optimization will not be corrected.

[0037] Specifically, obtaining the triggering status of compliance adjustment, heart rate update, and status optimization refers to obtaining data on whether compliance adjustment, heart rate update, and status optimization are triggered. In this embodiment, the triggering status of compliance adjustment, heart rate update, and status optimization is obtained through task logs.

[0038] Specifically, the exercise feedback module acquires the number of days of delay, determines the delay status based on the number of days of delay, and reconstructs the exercise rehabilitation plan based on the delay status, specifically including: The number of delayed days A is obtained, and the delay percentage C is calculated based on the number of delayed days A and the initial number of days B of the exercise rehabilitation plan. The delay percentage C is then compared with the preset delay percentage C0. Based on the comparison results, the delay situation is judged, and the exercise rehabilitation plan is reconstructed based on the judgment results. When C < C0, the motion feedback module determines that the delay is not serious and does not reconstruct the exercise rehabilitation plan; When C≥C0, the motion feedback module determines that the delay is severe and reconstructs the exercise rehabilitation plan. The reconstruction includes sending the target user data to medical staff, who then formulate the exercise rehabilitation plan.

[0039] Specifically, the preset delay percentage refers to a preset value used to judge the delay situation. The standard recovery period after PCI is 8 weeks (56 days). When the delay exceeds 14 days (56 × 25%), it means that the patient's rehabilitation compliance or physiological recovery ability has been deviating from the expected trajectory for more than 2 weeks. At this time, the simple extension plan can no longer solve the problem, and the patient may experience unrecognized complications (such as in-stent restenosis, deterioration of cardiac function). Severe psychological disorders (depression, post-traumatic stress) leading to behavioral inhibition, collapse of family / social support systems (lack of supervision, economic pressure interrupting rehabilitation), and the dual risks of continued automatic adjustment potentially leading to insufficient rehabilitation or overtraining necessitate a reassessment of the pathophysiological state by medical staff and a reconstruction of the exercise rehabilitation program. Therefore, C0=25% is set. The aforementioned delay situation refers to the severity of the delay as determined by comparing the delay percentage with the preset delay percentage. The delay situation includes severe delay and non-severe delay. In this embodiment, target user data is sent to medical staff via wireless signal.

[0040] Specifically, the motion feedback module integrates five levels of closed-loop control logic: compliance adjustment, heart rate update, state optimization, co-triggered correction, and delayed reconstruction. The co-triggered correction mechanism adopts a priority arbitration strategy of heart rate update > compliance adjustment > state optimization to avoid scheme conflicts and over-adjustment when multiple rules are triggered in parallel. It solves the technical problem of poor balance between adaptive adjustment capability and medical safety, and realizes the coupling of the system's daily automatic adjustment response time with the rigid safety circuit breaker of the 25% delay ratio threshold for manual intervention. While improving the adaptability of the scheme, it reduces the incidence of risk events of insufficient rehabilitation and overtraining by 65%, achieving a dynamic balance between adaptive adjustment and medical safety.

[0041] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A dynamic generation system for personalized exercise rehabilitation programs after PCI, characterized in that, The system includes: The user data collection module is used to collect target user data, which includes the target user's occupation type, surgical report, and risk assessment level. The exercise plan generation module is used to obtain exercise needs based on the user's occupation type, and also to generate exercise rehabilitation plans based on exercise needs, surgical reports, and risk assessment levels. The exercise program push module is used to push exercise rehabilitation programs to the target user's terminal; The exercise feedback module is used to acquire exercise data from the target user's terminal, calculate an objective compliance index based on the exercise data, judge the user's compliance status based on the objective compliance index, adjust the exercise rehabilitation plan based on the judgment results, judge the user's training load based on the heart rate in the exercise data, update the heart rate of the exercise rehabilitation plan based on the judgment results, judge the user's recovery status based on the training duration and heart rate in the exercise plan, optimize the status of the exercise rehabilitation plan based on the judgment results, acquire common triggering conditions, correct the compliance adjustment, heart rate update and status optimization process based on common triggering conditions, acquire the number of delay days, judge the delay status based on the number of delay days, and reconstruct the exercise rehabilitation plan based on the delay status.

2. The dynamic generation system for personalized exercise rehabilitation programs after PCI according to claim 1, characterized in that, The exercise plan generation module obtains the user's exercise needs based on their occupation type, specifically including: When a user's occupation is manual laborer, their exercise needs are: primarily to improve cardiorespiratory endurance and core stability, with a focus on aerobic training and resistance training that simulates work postures; When a user's occupation is a sedentary office worker, their exercise needs are: mainly to improve circulation and correct posture, with a focus on dynamic stretching, intermittent activity reminders, and spinal stability. When a user's occupation type is shift worker, their exercise needs are: mainly to regulate their biological clock and combat fatigue, with a focus on light-synchronized exercise and fatigue-resistant training; When a user's profession is a high-pressure decision-maker, their exercise needs are: mainly focused on stress hormone management and neuroregulation, with a focus on aerobic training and mindfulness exercises.

3. The dynamic generation system for personalized exercise rehabilitation programs after PCI according to claim 1, characterized in that, The exercise program generation module generates an exercise rehabilitation program based on exercise needs, surgical reports, and risk assessment levels, specifically including: The exercise needs, surgical reports, and risk assessment levels are input as user health data into the exercise plan generation model to obtain the exercise rehabilitation plan output by the exercise plan generation model. The exercise plan generation module constructs the exercise plan generation model using the exercise plan generation model construction method.

4. The dynamic generation system for personalized exercise rehabilitation programs after PCI according to claim 1, characterized in that, The motion scheme generation model construction method includes: The system processes historical user health data to develop exercise rehabilitation plans, identifies the characteristics of exercise rehabilitation needs in user health environment data, extracts these characteristics, and then uses exercise rehabilitation plan model learning to generate an exercise plan generation model with exercise rehabilitation plan output. The extraction of sports rehabilitation needs features involves extracting the multidimensional matrix data of user health environment data after sports type analysis and sports parameter optimization, and then performing sports rehabilitation program model calculations. This includes: extracting sports type features, sports load features, and sports parameter configuration features from the sports rehabilitation needs features, as well as extracting sports rehabilitation quality information from the sports rehabilitation needs features. Specifically, the extraction of exercise type characteristics from the exercise rehabilitation needs features is achieved through the output set of three pathways constituted by the exercise type analysis layer, including, from left to right: the first pathway is the occupational type analysis layer plus the work mode analysis layer; the second pathway is the risk stratification analysis layer plus the contraindicated movement analysis layer; and the third pathway is the functional state analysis layer plus the rehabilitation stage analysis layer. The extraction of exercise load characteristics from the exercise rehabilitation needs features is achieved through the output set of three pathways constituted by the exercise load configuration layer, including, from left to right: the first pathway is the exercise intensity configuration layer plus the intensity progression rule layer; the second pathway is the exercise frequency configuration layer plus the frequency cycle optimization layer; and the third pathway is the single session duration configuration layer plus the inter-session rest configuration layer. The extraction of exercise parameter configuration characteristics from the exercise rehabilitation needs features is achieved through the output set of three pathways constituted by the exercise parameter optimization layer, including, from left to right: the first pathway is the target heart rate parameter layer plus the RPE parameter layer; the second pathway is the contraindicated movement filtering layer plus the environmental compensation layer; and the third pathway is the exercise form selection layer plus the exercise sequence arrangement layer. The output of the exercise rehabilitation plan is implemented by the exercise rehabilitation plan output module. This module dynamically adjusts the contribution of each input feature through a learnable exercise rehabilitation plan generation network, and then performs exercise rehabilitation plan generation calculation to form a unified exercise rehabilitation plan output. The extraction of exercise rehabilitation quality information from the exercise rehabilitation demand features is realized through the exercise rehabilitation quality assessment mechanism. The number of parallel computing lines of this mechanism is greater than the number of model layer computing lines for extracting exercise type features, exercise load features and exercise parameter configuration features from the exercise rehabilitation demand features.

5. The dynamic generation system for personalized exercise rehabilitation programs after PCI according to claim 1, characterized in that, The exercise program push module pushes the exercise rehabilitation program to the target user's terminal in real time via the WebSocket long connection protocol.

6. The dynamic generation system for personalized exercise rehabilitation programs after PCI according to claim 1, characterized in that, The motion feedback module acquires motion data from the target user's terminal, calculates an objective compliance index based on the motion data, assesses user compliance based on the objective compliance index, and adjusts the exercise rehabilitation plan based on the assessment results. Specifically, this includes: The motion feedback module acquires motion data by receiving motion data fed back from the target user terminal; The objective compliance index K is calculated based on the actual exercise duration Ts from the exercise data and the planned exercise duration Tj in the exercise rehabilitation program, with K = Ts / Tj. The objective compliance index K is then compared with a preset objective compliance index K0. Based on the comparison results, user compliance is assessed, and adjustments are made to the exercise rehabilitation program accordingly. When K≥K0, the motion feedback module determines that the user's compliance is good and does not adjust the exercise rehabilitation plan. When K < K0, the exercise feedback module determines that the user's compliance is poor and adjusts the exercise rehabilitation plan accordingly. The compliance adjustment includes replacing the exercise plan for the next day with the exercise plan for the previous day and extending the total cycle of the exercise plan by 1 day.

7. The dynamic generation system for personalized exercise rehabilitation programs after PCI according to claim 6, characterized in that, The exercise feedback module determines the user's training load based on the heart rate in the exercise data, and updates the heart rate in the exercise rehabilitation plan according to the determination result, specifically including: When the real-time heart rate collected in the exercise data exceeds the heart rate threshold in the exercise intensity for 5 consecutive minutes, the exercise feedback module determines that the user's training load is overloaded and updates the heart rate of the exercise rehabilitation plan. The heart rate update includes: reverting the exercise plan for the next day to the exercise plan of the previous two days and extending the total cycle of the exercise plan by 2 days. If the real-time heart rate collected in the exercise data does not exceed the heart rate threshold in the exercise intensity for 5 consecutive minutes, the exercise feedback module determines that the user's training load is normal and does not update the heart rate in the exercise rehabilitation plan. When the number of times the heart rate exceeds the threshold reaches 3 times in a single day, the exercise feedback module determines that the user's training load is overloaded and updates the heart rate of the exercise rehabilitation plan. The heart rate update includes: reverting the exercise plan for the next day to the exercise plan of the previous two days and extending the total cycle of the exercise plan by 2 days. When the number of times the heart rate exceeds the threshold is less than 3 times in a single day, the exercise feedback module determines that the user's training load is normal and does not update the heart rate of the exercise rehabilitation plan.

8. The dynamic generation system for personalized exercise rehabilitation programs after PCI according to claim 7, characterized in that, The exercise feedback module assesses the user's recovery status based on the training duration and heart rate in the exercise plan, and optimizes the exercise rehabilitation plan based on the assessment results, specifically including: When a user's cumulative daily exercise time exceeds 125% of the duration in the exercise plan, and the real-time heart rate does not exceed the heart rate threshold throughout the entire process, the exercise feedback module determines that the user's recovery status is excellent and optimizes the exercise rehabilitation plan. The optimization includes: advancing the next day's exercise plan to the next day's exercise plan in advance, and shortening the total cycle of the exercise plan by 1 day. When a user's cumulative daily exercise time exceeds 125% of the duration in the exercise plan and the real-time heart rate exceeds the heart rate threshold, the exercise feedback module determines that the user's recovery status is "general recovery" and does not optimize the exercise rehabilitation plan. When a user's cumulative daily exercise time is less than 125% of the time in the exercise plan, and the real-time heart rate does not exceed the heart rate threshold throughout the exercise, the exercise feedback module determines that the user's recovery status is "general recovery" and does not optimize the exercise rehabilitation plan. When a user's cumulative daily exercise time is less than 125% of the time in the exercise plan, and the real-time heart rate exceeds the heart rate threshold, the exercise feedback module determines that the user's recovery status is "general recovery" and does not optimize the exercise rehabilitation plan.

9. The dynamic generation system for personalized exercise rehabilitation programs after PCI according to claim 8, characterized in that, The motion feedback module acquires co-triggered conditions and corrects the compliance adjustment, heart rate update, and status optimization processes based on these conditions, specifically including: The system acquires the triggering conditions for compliance adjustment, heart rate update, and state optimization, and obtains common triggering conditions. When the common triggering conditions satisfy two or more of the three triggering conditions, the process of compliance adjustment, heart rate update, and state optimization is corrected. The correction includes: the motion feedback module executes a unique triggering strategy according to the triggering priority order. The displacement triggering strategy means that only one of compliance adjustment, heart rate update, and state optimization is triggered. The triggering priority is: heart rate update > compliance adjustment > state optimization. When two or more of the following triggering conditions are not met: compliance adjustment, heart rate update, and state optimization, the process of compliance adjustment, heart rate update, and state optimization will not be corrected.

10. The dynamic generation system for personalized exercise rehabilitation programs after PCI according to claim 1, characterized in that, The exercise feedback module acquires the number of days of delay, assesses the delay based on the number of days, and reconstructs the exercise rehabilitation plan based on the delay situation, specifically including: The number of delayed days A is obtained, and the delay percentage C is calculated based on the number of delayed days A and the initial number of days B of the exercise rehabilitation plan. The delay percentage C is then compared with the preset delay percentage C0. Based on the comparison results, the delay situation is judged, and the exercise rehabilitation plan is reconstructed based on the judgment results. When C < C0, the motion feedback module determines that the delay is not serious and does not reconstruct the exercise rehabilitation plan; When C≥C0, the motion feedback module determines that the delay is severe and reconstructs the exercise rehabilitation plan. The reconstruction includes sending the target user data to medical staff, who then formulate the exercise rehabilitation plan.

Images