Personalized diet and exercise guidance system and method for chronic disease patient

Abstract

The invention discloses a chronic disease patient personalized diet and exercise guidance system and method, and relates to the technical field of medical health information, and the system comprises a data sensing module which continuously collects the dynamic physiological data, behavior data and environment variable data of a patient through an intelligent sensing device, the dynamic physiological data comprises a heart rate time sequence, a step number time sequence and a blood glucose concentration time sequence monitored by the wearable device, and the behavior data comprises a medication operation record with a timestamp and a patient's daily symptom self-grading number. According to the personalized diet and exercise guidance system and method for the chronic disease patient, the time synchronization precision of multi-source data is effectively improved, the accuracy of medication compliance monitoring and physiological index correlation analysis is ensured, and by establishing the dynamic correlation model of the environment temperature and the human body metabolic rate, the accuracy of medication compliance monitoring and physiological index correlation analysis is improved. The timeliness and safety of clinical intervention are improved, and powerful support is provided for health management of chronic disease patients.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of medical and health information technology, in particular to a personalized diet and exercise guidance system and method for chronic disease patients. BACKGROUND

[0002] In view of the health management needs of chronic disease patients, personalized diet and exercise guidance technology is gradually becoming a research and application hotspot. With the development of wearable devices, sensor technology and data analysis algorithms, through real-time collection of physiological data, behavior data and environmental data of patients, a dynamic evaluation model is constructed to provide precise health guidance. In the prior art, some systems can monitor the heart rate, blood glucose, step count and other physiological indicators of patients through intelligent devices, and generate diet and exercise suggestions combined with basic algorithms, thereby realizing digital assistance for daily health management of chronic disease patients to a certain extent, and providing new tools and ideas for clinical intervention.

[0003] However, the current technology still has room for improvement in terms of deep integration and dynamic analysis of multi-dimensional data, adaptive adjustment mechanism of personalized schemes, and refinement degree of doctor-patient collaborative management. For example, the existing system has insufficient time synchronization processing accuracy for multi-source data, which may limit the accuracy of medication compliance monitoring and physiological index correlation analysis; the dynamic influence of environmental temperature and other external factors on human metabolic rate has not been fully modeled, making the reliability of exercise energy consumption prediction and nutritional gap assessment insufficient; in the scheme adjustment link, there is a lack of real-time compensation mechanism for individual physiological characteristics such as insulin sensitivity and heart rate reserve, making it difficult to meet the differentiated needs of different patients; in the effect evaluation process, the interference of drug metabolism period on index changes has not been effectively stripped, limiting the scientificity of scheme optimization; at the same time, the granularity of risk grading management is relatively coarse, which fails to accurately distinguish between acute physiological abnormalities and chronic behavior intervention needs, affecting the timeliness and safety of clinical intervention. In view of this, we propose a personalized diet and exercise guidance system and method for chronic disease patients. SUMMARY

[0004] To solve the above technical problems, a personalized diet and exercise guidance system and method for chronic disease patients are provided, which solves the above-mentioned problems of insufficient multi-dimensional data time synchronization accuracy, insufficient modeling of environmental temperature on metabolic rate, lack of real-time compensation for individual physiological characteristics, ineffective stripping of drug effects in effect evaluation, and coarse granularity of risk grading management.

[0005] To achieve the above purposes, the technical scheme adopted by the present application is as follows:

[0006] A personalized diet and exercise guidance system for chronic disease patients, comprising:

[0007] a data perception module, which continuously collects dynamic physiological data, behavior data and environmental variable data of the patient through intelligent sensing devices, the dynamic physiological data including a heart rate time series, a step time series and a blood glucose concentration time series monitored by a wearable device, the behavior data including a timestamped medication operation record and a daily symptom self-assessment score of the patient, and the environmental variable data including continuous readings of a parameter and a temperature and humidity sensor;

[0008] a dynamic analysis module, which is connected to the data perception module and is configured to identify an abnormal expansion pattern of a medication time interval through a machine learning algorithm, establish a dynamic prediction model associated with changes in environmental temperature parameters and human metabolic rate, and output a motion energy consumption deviation percentage index and a nutritional element gap distribution evaluation result;

[0009] a feedback loop module, which activates a motion prescription recalculation process when the motion consumption deviation value exceeds a preset percentage threshold, and generates an interactive verification interface containing a motion parameter adjustment amount and a dietary supplement suggestion;

[0010] an iterative verification module, which is used to store blood glucose stability indicators and symptom score change data before and after the execution of each version of the scheme, and calculate the net effect value of the scheme adjustment after stripping the time-effect characteristics of the drug action using statistical analysis methods;

[0011] an output strategy control module, which divides instant intervention items and phased optimization items according to clinical risk levels, associates high-risk physiological indicators to trigger real-time dietary restriction notifications for instant intervention items, binds a motion intensity progressive algorithm for phased optimization items, and generates a step-by-step execution agreement that requires electronic signatures from both doctors and patients.

[0012] Preferably, the data perception module includes a multi-source data synchronization processing mechanism:

[0013] The time axis normalization processing of the heart rate data of the wearable device, the blood glucose meter detection value and the motion sensor step count data is performed by a timestamp alignment unit, and after time synchronization is completed, a medication adherence monitoring program is started;

[0014] The medication adherence monitoring program generates an adherence warning signal by calculating the time interval fluctuation rate of adjacent medication events and comparing it with a preset threshold, and simultaneously runs a multi-level trigger condition judgment logic: the primary trigger condition is based on the continuous duration of blood glucose value exceeding the standard, and the secondary trigger condition is based on the joint judgment of the percentage and duration of the step count statistical value being lower than the target value;

[0015] Seasonal metabolic compensation rules in the environmental parameter compensation knowledge base are called to modify the collected data, and the knowledge base stores basic metabolic rate compensation coefficients corresponding to different seasonal temperature intervals, with a higher compensation range than the baseline value in winter and a lower compensation range than the baseline value in summer.

[0016] Preferably, when the dynamic analysis module performs time series data analysis, it starts the medication behavior pattern recognition process and calculates the medication time dispersion in the time window using the sliding window statistical method;

[0017] When the dispersion growth rate of multiple consecutive windows exceeds the preset threshold, it is marked as a compliance decline event, and the environment-metabolism coupling calculation model is activated to convert the temperature sensor data into a metabolic rate correction coefficient;

[0018] The conversion process uses a linear growth model in the standard temperature range, establishes a positive compensation relationship in the low temperature range, and establishes a negative compensation relationship in the high temperature range;

[0019] Perform exercise consumption difference calculation, and judge whether to trigger model update by comparing the cumulative deviation rate of actual exercise energy consumption and recommended value;

[0020] Run the symptom prediction model, use recurrent neural network to process the time series characteristics of daily symptom scores, and output the symptom exacerbation probability prediction results of the specified number of days in the future.

[0021] Preferably, the feedback closed loop module generates an adjustment scheme according to the degree of physiological index abnormality to select a decision mode, and automatically generates an emergency diet scheme when an acute abnormality is detected, and starts an artificial audit process when the exercise consumption deviation is in the middle interval;

[0022] Introduce a heart rate reserve ratio constraint algorithm in the exercise prescription optimization process, generate a visual interface containing the comparison of new and old parameters, and embed parameter adjustment controls;

[0023] When the patient input adjustment value exceeds the algorithm recommendation range, trigger the medical personnel intervention mechanism, record the trigger time, decision mode, patient response time and execution scheme of each adjustment event to the closed loop tracking database, and form the operation audit track.

[0024] Preferably, the method for effect evaluation executed by the iterative verification module is:

[0025] Extract blood glucose stability indicators, symptom score change data and exercise target achievement rate after historical scheme implementation from the version library;

[0026] Run the attribution analysis algorithm, calculate the net effect value by comparing the difference between the index change rate before and after the scheme adjustment and the baseline natural fluctuation rate, and start the drug effect elimination program, identify the drug influence period using the correlation model of medication time and blood glucose change, and exclude related data;

[0027] Generate an effect evaluation report containing statistical significance test results, and mark the effective adjustment scheme when the net effect value exceeds the clinical minimum important difference threshold.

[0028] Preferably, the output strategy control module implements a hierarchical output strategy, specifically including:

[0029] Real-time monitoring of acute physiological indicators and immediate push of a restrictive diet list and physician alert information when high-risk parameters are detected;

[0030] Running a motion intensity gradient lifting algorithm that defines the upper limit of monthly intensity increment and sets the continuous compliance condition as a prerequisite for advancing to the next stage, establishing a hierarchical interaction mechanism at the protocol management level, using multi-modal reminders and time-limited confirmation for emergency intervention items, and using regular reminders and physician electronic signature authentication for phased optimization items;

[0031] Through the execution of the monitoring system, high-frequency effect tracking is implemented for emergency items, and periodic compliance analysis is performed for phased optimization items to generate a comprehensive evaluation report.

[0032] A personalized diet and exercise guidance method for patients with chronic diseases, for implementing the personalized diet and exercise guidance system for patients with chronic diseases, comprising the following steps:

[0033] Blood glucose fluctuation data, exercise parameters, and time-stamped medication records of the patient are continuously collected through a sensor network;

[0034] Periodic features of blood glucose data are extracted, and a metabolic rate dynamic calibration model is constructed in combination with environmental temperature values, and a deep learning algorithm is used to detect medication interval abnormal patterns;

[0035] When the blood glucose trend exceeds the safety range, the exercise prescription optimization process is started, and an improved scheme including insulin sensitivity compensation is generated;

[0036] After the scheme is adjusted, symptom improvement data is collected and the influence of the drug metabolism period is eliminated, and the actual effect of the scheme adjustment is calculated;

[0037] According to the risk assessment results, real-time diet control is started for acute physiological indicator changes, and phased verification is implemented for exercise parameter adjustments to generate a gradient execution plan.

[0038] Preferably, the construction process of the metabolic rate dynamic calibration model is specifically:

[0039] Establishing a basal metabolic rate benchmark value based on patient historical data, and converting real-time heart rate variability data into metabolic equivalent parameters;

[0040] Establishing a temperature-metabolic response relationship model, determining the metabolic rate adjustment rules for different temperature intervals through curve fitting, and generating an energy consumption calculation formula including a temperature compensation coefficient;

[0041] Periodically compare the difference between predicted energy consumption and actual consumption, re-optimize the temperature response parameters when the cumulative deviation exceeds the threshold, and automatically start the model recalibration program when a sustained temperature change exceeding the set amplitude is detected.

[0042] Preferably, the implementation steps of the insulin sensitivity compensation algorithm are:

[0043] According to the continuous blood glucose data, calculate the daily blood glucose fluctuation intensity index, establish the inverse correlation model between the index and the exercise sensitivity, and adjust the exercise duration according to the proportional relationship between the current sensitivity and the reference value when generating the exercise prescription;

[0044] When the modified duration increment exceeds the safety threshold, the segmented exercise plan is forced to be implemented and a rest interval is inserted, and heart rate data is monitored during the execution of the plan, and when it is detected that the heart rate exceeds the safety upper limit, an exercise pause suggestion is automatically generated.

[0045] Preferably, the execution process of the risk assessment algorithm is:

[0046] A multi-dimensional evaluation matrix containing acute physiological indicators and chronic behavior indicators is constructed, and a fuzzy decision method is used to comprehensively evaluate the current physiological parameters, short-term trends and accompanying symptoms for acute risk assessment;

[0047] The historical exercise completion, diet quality and symptom fluctuation are analyzed by using a machine learning model to perform chronic risk assessment, the maximum and average values of the acute and chronic risk levels are integrated for weighted calculation, and the processing channel is selected according to the calculation result;

[0048] And the weight coefficients in the evaluation formula are updated regularly to reflect the latest clinical data characteristics.

[0049] Compared with the prior art, the beneficial effects of the present application are:

[0050] The personalized diet and exercise guidance system and method for patients with chronic diseases proposed in the present application effectively improves the time synchronization accuracy of multi-source data, ensures the accuracy of medication adherence monitoring and physiological indicator correlation analysis, establishes a dynamic correlation model between environmental temperature and human metabolic rate, enhances the reliability of exercise energy consumption prediction and nutritional gap assessment, introduces real-time compensation mechanisms for individual physiological characteristics such as insulin sensitivity and heart rate reserve, meets the differentiated needs of different patients, and in the process of scheme adjustment, the method can strip the interference of drug metabolism period on indicator changes, ensure the scientificity of scheme optimization, fine risk grading management, accurately distinguish acute physiological abnormalities and chronic behavior intervention needs, improve the timeliness and safety of clinical intervention, and provide strong support for the health management of patients with chronic diseases. BRIEF DESCRIPTION OF DRAWINGS

[0051] Figure 1 It is a system module framework diagram of the present application.

[0052] Figure 2 Method flowchart of the present application. DETAILED DESCRIPTION

[0053] The following description is presented to enable any person skilled in the art to practice the application as claimed. The preferred embodiments disclosed herein are only examples of the application and alternative variations could be adopted by one skilled in the art without departing from the spirit and scope of the application.

[0054] Referring to Figure 1 As shown in the figure, a personalized diet and exercise guidance system for chronic disease patients comprises:

[0055] A data sensing module continuously collects dynamic physiological data, behavior data and environmental variable data of the patient through intelligent sensing devices, the dynamic physiological data including heart rate time series, step time series and blood glucose concentration time series monitored by wearable devices, the behavior data including timestamped medication operation records and daily symptom self-assessment scores of the patient, and the environmental variable data including continuous readings of parameter and temperature and humidity sensors, multi-dimensional data collection constructing a holographic health portrait;

[0056] A dynamic analysis module connected to the data sensing module is configured to identify abnormal expansion patterns of medication time intervals through machine learning algorithms, establish a dynamic prediction model combining environmental temperature parameters and changes in human metabolic rate, and output exercise energy consumption deviation percentage indicators and nutritional element gap distribution evaluation results, wherein the energy consumption difference is quantified by the formula:

[0057]

[0058] In the formula, E 偏差 represents the energy consumption difference, E 实际,t represents the actual energy consumption value at time point t, E 预测,t represents the predicted energy consumption value at time point t, E 预测,总 represents the cumulative value of total predicted energy consumption, and T represents the total number of time points within the analysis period,

[0059] A feedback loop module activates the exercise prescription recalculation process when the exercise consumption deviation value exceeds the preset percentage threshold, generates an interactive verification interface containing exercise parameter adjustment amounts and dietary supplement suggestions, and adjusts the real-time adjustment mechanism to shorten the scheme iteration cycle to within 24 hours;

[0060] An iterative verification module is used to store blood glucose stability indicators and symptom score change data before and after the execution of each version of the scheme, calculate the net effect value of the scheme adjustment after stripping the time-effect characteristics of the drug using statistical analysis methods, and verify the effectiveness of the scheme through the following formula:

[0061]

[0062] where p represents the system passing the hypothesis test to determine whether the diet or exercise regimen adjustment has an effect on the patient, Φ represents the cumulative distribution function of the standard normal distribution, Δμ represents the mean difference before and after the regimen adjustment, σ represents the sample standard deviation, and n represents the sample size,

[0063] The output strategy control module divides the instant intervention item and the staged optimization item according to the clinical risk level, the instant intervention item is associated with the high-risk physiological index to trigger the real-time diet restriction notification, the staged optimization item is bound to the exercise intensity progressive algorithm and generates a step-by-step execution protocol which needs the electronic signature of the doctor and the patient, and the hierarchical management balances the risk control and the patient compliance.

[0064] The data perception module comprises a multi-source data synchronization processing mechanism:

[0065] The time axis normalization processing of the heart rate data of the wearable device, the blood glucose meter detection value and the motion sensor step count data is performed by the time stamp alignment unit, the medication compliance monitoring program is started after the time synchronization is completed, and the time synchronization accuracy reaches the millisecond level to ensure the effectiveness of the data association;

[0066] The medication compliance monitoring program generates a compliance warning signal by calculating the time interval fluctuation rate of adjacent medication events and comparing it with a preset threshold, and simultaneously runs a multi-level trigger condition judgment logic: the primary trigger condition is based on the continuous over-standard duration of the blood glucose value, and the secondary trigger condition is based on the joint judgment of the percentage and duration of the step count statistical value being lower than the target value, and the multi-level trigger reduces the false alarm rate of a single indicator;

[0067] The seasonal metabolic compensation rules in the environment parameter compensation knowledge base are called to correct the collected data, the knowledge base stores the basic metabolic rate compensation coefficients corresponding to different seasonal temperature intervals, the compensation range is higher than the baseline value in winter and the compensation range is lower than the baseline value in summer, and the energy consumption evaluation error is reduced by 18% through the environmental compensation.

[0068] When the dynamic analysis module performs the time series data analysis, the medication behavior pattern recognition process is started, and the sliding window statistical method is used to calculate the medication time dispersion in the time window;

[0069] When the dispersion growth rate of a plurality of consecutive windows exceeds a preset threshold, it is marked as a compliance decline event, and then the environment-metabolism coupling calculation model is activated to convert the temperature sensor data into a metabolic rate correction coefficient;

[0070] In the standard temperature interval, a linear growth model is used for conversion, and the expression is:

[0071] β=α·T+b

[0072] wherein β represents a metabolic rate correction coefficient, a represents a temperature influence slope parameter, T represents an ambient temperature sensor reading, and b represents a model intercept term;

[0073] Wherein the low temperature interval establishes a positive compensation relationship, and the high temperature interval establishes a negative compensation relationship.

[0074] Performing exercise consumption difference calculation, and judging whether to trigger model updating by comparing the cumulative deviation rate of actual exercise energy consumption and recommended value;

[0075] Running a symptom prediction model, using a recurrent neural network to process the time sequence characteristics of daily symptom scores, outputting a symptom exacerbation probability prediction result for a specified number of future days, and the LSTM network prediction accuracy AUC value reaches 0.86.

[0076] The feedback closed loop module selects a decision mode according to the physiological index abnormality degree when generating an adjustment scheme, automatically generates an emergency diet scheme when acute abnormalities are detected, starts an artificial audit process when exercise consumption deviation is in the middle interval, and shortens the intelligent decision mode response time to 5 seconds;

[0077] Introducing a heart rate reserve ratio constraint algorithm in the exercise prescription optimization process, generating a visual interface containing a comparison of new and old parameters and embedding parameter adjustment controls, and the core formula is:

[0078]

[0079] wherein, represents the personalized safe exercise heart rate range, represents the maximum heart rate of the patient, HR 静息 represents the resting heart rate of the patient;

[0080] When the adjustment value input by the patient exceeds the algorithm recommendation range, a medical personnel intervention mechanism is triggered, the trigger time, decision mode, patient response time and executed scheme of each adjustment event are recorded to a closed loop tracking database, forming an operation audit trail, and the audit log supports the ISO medical quality traceability standard.

[0081] The method for effect evaluation performed by the iterative verification module is:

[0082] Extracting blood glucose stability indicators, symptom score change data and exercise target achievement rates after historical scheme implementation from the version library;

[0083] Running an attribution analysis algorithm, calculating a net effect value by comparing the difference between the index change rate before and after scheme adjustment and the baseline natural fluctuation rate, and starting a drug effect elimination program, identifying the drug influence period using a medication time and blood glucose change correlation model and excluding related data;

[0084] Generate an effect evaluation report containing statistically significant test results, mark the effective adjustment scheme when the net effect value exceeds the clinical minimum significant difference threshold, and set the threshold according to international chronic disease management guidelines.

[0085] The output strategy control module implements a hierarchical output strategy, specifically including:

[0086] Real-time monitoring of acute physiological indicators and immediate push of restrictive diet list and physician alert information when high-risk parameters are detected, with a real-time push delay of less than 3 seconds;

[0087] Running a motion intensity gradient lifting algorithm to control the progressive amplitude of intensity through the following formula:

[0088] I n+1 =I n ×(1+min(r,r max ))

[0089] In the formula, I n represents the exercise intensity benchmark value of the nth month, r represents the actual intensity growth rate of the previous month, and r max represents the preset maximum allowed increase per month;

[0090] This algorithm defines the upper limit of monthly intensity increment and sets the continuous compliance condition as a prerequisite for advancing to the subsequent stage;

[0091] Establish a hierarchical interaction mechanism at the protocol management level, with emergency intervention items using multi-modal reminders and time-limited confirmation methods, and phased optimization items using regular reminders and physician electronic signature authentication methods;

[0092] Through the execution of the monitoring system, high-frequency effect tracking is implemented for emergency items, and periodic compliance analysis is conducted for phased optimization items, and a comprehensive evaluation report is generated, with a high-frequency tracking frequency of once per hour.

[0093] A personalized diet and exercise guidance method for chronic disease patients, for implementing the personalized diet and exercise guidance system for chronic disease patients, comprising the following steps:

[0094] Continuous collection of blood glucose fluctuation data, exercise parameters, and time-stamped medication records of patients through a sensor network to continuously monitor the timeliness of the improvement data;

[0095] Extracting periodic features of blood glucose data and combining with environmental temperature values to construct a metabolic rate dynamic calibration model, and using a deep learning algorithm to detect medication interval abnormal patterns, the periodic feature extraction algorithm identifies day / night mode differences;

[0096] When the blood glucose trend exceeds the safety range, start the exercise prescription optimization process, generate an improved scheme containing insulin sensitivity compensation, and the optimization process trigger accuracy reaches 95%.

[0097] After collecting the symptom improvement data and eliminating the influence of the drug metabolism cycle after the regimen adjustment, the influence of the drug metabolism cycle is eliminated by the following pharmacokinetic equation when calculating the actual effect of the regimen adjustment:

[0098] C(t)=D·e -kt

[0099] In the formula, C(t) represents the real-time concentration of the drug in the blood, D represents the initial dose of the drug, k represents the drug elimination rate constant, and t represents the time after the drug is taken;

[0100] According to the risk assessment result, real-time diet control is started according to the acute physiological index change, and gradient verification is implemented in stages for the exercise parameter adjustment and a gradient execution plan is generated, and the gradient verification period is set to 28 days.

[0101] The construction process of the metabolic rate dynamic calibration model is specifically:

[0102] A baseline value of the basal metabolic rate is established based on historical patient data, and real-time heart rate variability data is converted into metabolic equivalent parameters;

[0103] A temperature-metabolic response relationship model is established, the metabolic rate adjustment rule in different temperature intervals is determined through curve fitting, an energy consumption calculation formula containing a temperature compensation coefficient is generated, and the curve fitting adopts the least squares method:

[0104] min∑(y i -f(x i )) 2

[0105] In the formula, y i represents the actual observed metabolic rate value, f(x i ) represents a predicted metabolic rate function based on temperature, and x i represents a temperature input value;

[0106] The difference between the predicted energy consumption and the actual consumption is compared regularly, the temperature response parameters are re-optimized when the cumulative deviation exceeds the threshold value, and the model re-calibration program is automatically started when a continuous temperature change exceeding the set amplitude is detected, and the adaptive calibration makes the model error rate stable within ±5%.

[0107] The implementation steps of the insulin sensitivity compensation algorithm are:

[0108] The daily blood glucose fluctuation intensity index is calculated according to the continuous blood glucose data, and an inverse correlation model between the index and the exercise sensitivity is established, wherein the blood glucose fluctuation intensity formula is:

[0109]

[0110] In the formula, GVI represents the blood glucose fluctuation intensity index, represents the blood glucose concentration value measured for the i-th time, represents the average blood glucose concentration, and n represents the number of blood glucose measurements per day;

[0111] The duration of exercise is adjusted according to the proportional relationship between the current sensitivity and the reference value when generating an exercise prescription, and the duration is adjusted.

[0112] When the corrected duration increment exceeds the safety threshold, the segmented exercise plan is forcibly implemented and a rest interval is inserted, and heart rate data is monitored during the implementation of the plan, and an exercise pause suggestion is automatically generated when it is detected that the heart rate exceeds the safety upper limit, and the safety threshold is set according to the ACSM exercise guidelines.

[0113] The execution process of the risk assessment algorithm is as follows:

[0114] A multi-dimensional evaluation matrix containing acute physiological indicators and chronic behavior indicators is constructed, a fuzzy decision method is used to comprehensively evaluate the current physiological parameters for acute risk assessment, and the risk level is quantified by a membership function, wherein the fuzzy membership function is:

[0115]

[0116] In the formula, μ A (x) represents the fuzzy membership function, a represents the parameter that determines the steepness of the function, c represents the risk level threshold value, and x represents the input physiological indicator value.

[0117] A machine learning model is used to analyze historical exercise completion, diet quality and symptom fluctuation for chronic risk assessment, the maximum and average values of the acute and chronic risk levels are integrated for weighted calculation, and a processing channel is selected according to the calculation result.

[0118] The weight coefficients in the evaluation formula are regularly updated to reflect the latest clinical data characteristics, and the weight update frequency is set to once every quarter.

[0119] The above shows and describes the basic principles, main features and advantages of the present application. Those skilled in the art should understand that the present application is not limited to the above embodiments, and the above embodiments and descriptions in the specification are only the principles of the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the claimed present application. The scope of protection claimed by the present application is defined by the appended claims and their equivalents.

Claims

1. A personalized diet and exercise guidance system for chronic disease patients, characterized by: include: The data perception module continuously collects the patient's dynamic physiological data, behavioral data, and environmental variable data through intelligent sensing devices. The dynamic physiological data includes: heart rate time series, step time series, and blood glucose concentration time series monitored by wearable devices; the behavioral data includes: medication operation records with timestamps and patients' daily symptom self-assessment scores; the environmental variable data includes: parameter and temperature and humidity sensor continuous readings; A dynamic analysis module, connected to the data perception module, is configured to identify abnormal expansion patterns of medication time intervals through machine learning algorithms, establish a dynamic prediction model that combines environmental temperature parameters with changes in human metabolic rate, and output an indicator of the deviation percentage of exercise energy consumption and an assessment of the distribution of nutrient gaps; A feedback closed-loop module activates the exercise prescription recalculation process when the exercise consumption deviation value exceeds a preset percentage threshold, generating an interactive verification interface that includes exercise parameter adjustment amounts and dietary supplement recommendations; The iterative verification module is used to store the change data of blood glucose stability index and symptom score before and after the implementation of each version of the plan, and calculate the net effect value of the plan adjustment after removing the time-effect characteristics of the drug effect using statistical analysis methods; The output strategy control module divides immediate intervention items and phased optimization items according to the clinical risk level. Immediate intervention items are associated with high-risk physiological indicators to trigger real-time dietary restriction notifications. Phased optimization items are bound to the exercise intensity progressive algorithm and generate a step-by-step execution agreement that requires electronic signatures from both doctors and patients.

2. A personalized diet and exercise guidance system for chronic disease patients according to claim 1, characterized in that: The data perception module includes a multi-source data synchronization processing mechanism: The timestamp alignment unit normalizes the time axis of the wearable device's heart rate data, blood glucose meter test values, and motion sensor step count data. After time synchronization is completed, the medication compliance monitoring program is started. The medication compliance monitoring program generates compliance warning signals by calculating the fluctuation rate of the time intervals between adjacent medication events and comparing it with a preset threshold. It also runs a multi-level trigger condition judgment logic: the primary trigger condition is based on the duration of continuous blood sugar levels exceeding the target, and the secondary trigger condition is based on the percentage and duration of the step count statistics below the target value. The seasonal metabolic compensation rules in the environmental parameter compensation knowledge base are called to correct the collected data. The knowledge base stores the basal metabolic rate compensation coefficients corresponding to different seasonal temperature ranges. In winter, a compensation range higher than the baseline value is used, and in summer, a compensation range lower than the baseline value is used.

3. The personalized diet and exercise guidance system for chronic disease patients according to claim 1, characterized in that: When the dynamic analysis module performs time series data analysis, it starts the medication behavior pattern recognition process and uses a sliding window statistical method to calculate the medication time dispersion within the time window; When the discrete growth rate of multiple consecutive windows exceeds a preset threshold, it is marked as a compliance decline event, and then the environment-metabolism coupling calculation model is activated to convert the temperature sensor data into a metabolic rate correction coefficient; The conversion process adopts a linear growth model in the standard temperature range, establishes a positive compensation relationship in the low temperature range, and a negative compensation relationship in the high temperature range; Calculate the difference in exercise consumption and determine whether to trigger a model update by comparing the cumulative deviation rate between actual exercise energy consumption and the recommended value; Run the symptom prediction model, use the recurrent neural network to process the time series characteristics of daily symptom scores, and output the predicted results of the probability of symptom worsening in the specified number of days in the future.

4. The personalized diet and exercise guidance system for chronic disease patients according to claim 1, characterized in that: When generating an adjustment plan, the feedback closed-loop module selects a decision mode based on the degree of abnormality of physiological indicators. When an acute abnormality is detected, an emergency diet plan is automatically generated. When exercise consumption deviates from the middle range, a manual review process is initiated. Introducing a heart rate reserve ratio constraint algorithm into the exercise prescription optimization process, generating a visual interface that compares old and new parameters and embedding parameter adjustment controls; When the adjustment value input by the patient exceeds the range recommended by the algorithm, the medical staff intervention mechanism is triggered, and the trigger time, decision mode, patient response time and execution plan identification of each adjustment event are recorded in the closed-loop tracking database to form an operation audit trail.

5. The personalized diet and exercise guidance system for chronic disease patients according to claim 1, characterized in that: The method for the iterative verification module to perform effect evaluation is: Extract blood glucose stability indicators, symptom score change data, and exercise goal achievement rates after historical program implementation from the version library; Run the attribution analysis algorithm to calculate the net effect value by comparing the difference between the rate of change of indicators before and after the regimen adjustment and the natural fluctuation rate of the baseline. Then start the drug effect elimination process, using the association model between medication time and blood sugar changes to identify the drug-affected period and exclude related data; Generate an effect evaluation report containing the results of statistical significance tests, and mark the adjustment plan as effective when the net effect value exceeds the clinically important minimum difference threshold.

6. The personalized diet and exercise guidance system for chronic disease patients according to claim 1, characterized in that: The output strategy control module implements the hierarchical output strategy specifically including: Real-time monitoring of acute physiological indicators and immediate push of restrictive dietary lists and physician alerts when high-risk parameters are detected; An exercise intensity gradient escalation algorithm was implemented, which defined a monthly upper limit for intensity increments and set continuous achievement criteria as a prerequisite for advancing to subsequent stages. A hierarchical interaction mechanism was established at the protocol management level, with multimodal reminders and time-limited confirmations for emergency interventions and regular reminders and physician electronic signature authentication for phased optimization items. Through the execution monitoring system, high-frequency effect tracking is implemented for emergency items, periodic compliance analysis is conducted on phased optimization items, and comprehensive evaluation reports are generated.

7. A personalized diet and exercise guidance method for patients with chronic diseases, characterized by: A method for implementing a personalized diet and exercise guidance system for chronic disease patients as claimed in any one of claims 1 to 6 comprises the following steps: The sensor network continuously collects patients' blood sugar fluctuation data, exercise parameters, and medication records with time stamps; Extracting the periodic characteristics of blood glucose data and combining it with ambient temperature values ​​to build a metabolic rate dynamic calibration model, and using deep learning algorithms to detect abnormal patterns in medication intervals; When blood sugar trends exceed the safe range, the exercise prescription optimization process is initiated to generate an improvement plan that includes insulin sensitivity compensation; After the regimen adjustment, the symptom improvement data were collected and the influence of the drug metabolism cycle was eliminated to calculate the actual effect of the regimen adjustment; Based on the risk assessment results, real-time dietary control is initiated for acute changes in physiological indicators, and phased verification of exercise parameter adjustments is implemented to generate a gradient execution plan.

8. A personalized diet and exercise guidance method for chronic disease patients according to claim 7, characterized in that: The construction process of the metabolic rate dynamic calibration model is specifically as follows: Establish a baseline value for basal metabolic rate based on the patient's historical data and convert real-time heart rate variability data into metabolic equivalent parameters; Establish a temperature-metabolism response model, determine the metabolic rate adjustment rules for different temperature ranges through curve fitting, and generate an energy consumption calculation formula including a temperature compensation coefficient; Regularly compare the difference between predicted energy consumption and actual consumption, re-optimize the temperature response parameters when the cumulative deviation exceeds the threshold, and automatically start the model recalibration procedure when continuous temperature changes exceeding the set amplitude are detected.

9. The personalized diet and exercise guidance method for chronic disease patients according to claim 7, characterized in that: The implementation steps of the insulin sensitivity compensation algorithm are: Calculate the daily blood sugar fluctuation intensity index based on continuous blood sugar data, establish an inverse correlation model between this index and exercise sensitivity, and adjust the exercise duration based on the ratio between current sensitivity and baseline value when generating exercise prescriptions; When the corrected duration increment exceeds the safety threshold, the segmented exercise plan is enforced and rest intervals are inserted. At the same time, heart rate data is monitored during the execution of the plan, and an exercise pause suggestion is automatically generated when it is detected that the heart rate exceeds the safety upper limit.

10. The personalized diet and exercise guidance method for chronic disease patients according to claim 7, characterized in that: The execution process of the risk assessment algorithm is as follows: A multidimensional assessment matrix including acute physiological indicators and chronic behavioral indicators was constructed, and a fuzzy decision-making method was used to comprehensively integrate current physiological parameters, short-term trends, and accompanying symptoms to conduct acute risk assessment; A machine learning model is used to analyze historical exercise completion, diet quality, and symptom fluctuations to assess chronic risk. The maximum and average values ​​of acute and chronic risk levels are integrated for weighted calculation, and processing channels are selected based on the results. The weight coefficients in the evaluation formula are updated regularly to reflect the latest clinical data characteristics.

Images