Atrial fibrillation patient health management system under exercise tolerance parameter auxiliary feedback
By using the WD-RCNN model and moving average method in the health management system of atrial fibrillation patients, we can monitor and analyze the patient's exercise tolerance parameters in real time, and solve the problem that the existing system cannot continuously track and evaluate changes in exercise tolerance, realizing personalized health Management and remote intervention have improved the targeted treatment plan and the efficiency of medical resource utilization.
Patent Information
- Application Number
- CN202410969579.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-18
- Publication Date
- 2024-11-12
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing health management system for patients with atrial fibrillation lacks continuous tracking and dynamic evaluation of changes in exercise tolerance, which makes it difficult to adjust the treatment plan in a timely manner according to the patient's actual status.
A health management system with assisted feedback of motion tolerance parameters was designed. The wearable sensors collected physiological and motion parameters in real time, and data analysis was used to use WD-RCNN model and moving average method to predict the trend of motion tolerance changes, and through mobile applications and intelligence The device provides personalized health management plans and remote monitoring functions.
Accurate assessment and prediction of exercise tolerance levels of patients with atrial fibrillation have been achieved, helping medical staff to timely detect health risks, formulate personalized rehabilitation plans, improve the pertinence and effectiveness of treatment plans, and reduce medical costs.
Smart Images

Figure CN118924260A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of health management for patients with atrial fibrillation, and specifically relates to a health management system for patients with atrial fibrillation under the auxiliary feedback of exercise tolerance parameters. Background Art
[0002] With the continuous advancement of medical technology, atrial fibrillation (AF), as a common arrhythmia in clinical practice, has received increasing attention for its treatment and management methods. The current treatment of AF mainly focuses on controlling heart rate and rhythm, reducing the risk of stroke and its related complications and mortality. In addition, the existing health management systems for AF patients mainly focus on drug management and basic health monitoring, such as regular recording of indicators such as heart rate and blood pressure. However, these systems often ignore the importance of the patient's daily exercise tolerance in affecting AF.
[0003] Existing health management systems for patients with atrial fibrillation rely on simple physical activity questionnaires or self-reports when assessing patients' exercise tolerance, which lacks accuracy and real-time performance, making it difficult for the health management system to provide patients with targeted exercise and rehabilitation guidance. In addition, existing systems often lack continuous tracking and dynamic evaluation of changes in exercise tolerance in patients with atrial fibrillation, making it difficult to adjust treatment plans in a timely manner based on the patient's actual condition. Summary of the invention
[0004] The present invention provides a health management system for patients with atrial fibrillation under the auxiliary feedback of exercise tolerance parameters, which is used to solve the technical problems that the existing health management systems lack continuous tracking and dynamic evaluation of changes in exercise tolerance of patients with atrial fibrillation, and cannot predict the trend of changes in exercise tolerance, thereby making it difficult to adjust the treatment plan in time according to the actual condition of the patient.
[0005] In order to achieve the above-mentioned purpose, the technical solution adopted by the present invention is as follows: a health management system for patients with atrial fibrillation under the auxiliary feedback of exercise tolerance parameters, comprising:
[0006] Exercise tolerance monitoring module: used to collect the patient's physiological parameters in real time through wearable sensors and obtain the patient's exercise parameters using accelerometers and gyroscope sensors. The physiological parameters include the patient's heart rate H, blood pressure P, and blood oxygen saturation β; the exercise parameters include exercise intensity ν and exercise mode M. Physiological parameters and exercise parameters are collectively referred to as exercise tolerance data;
[0007] Data processing module: used for data cleaning, data standardization and data normalization of the exercise tolerance data;
[0008] Data analysis module: perform WD-RCNN model evaluation on the exercise tolerance data, and predict the change trend of the patient's exercise tolerance level in the future by analyzing the patient's historical exercise tolerance data and using the moving average method;
[0009] Feedback and health management module: formulate appropriate exercise plans according to the patient's exercise tolerance level, and push personalized health management plans and precautions to patients and medical staff in real time through mobile applications or smart devices;
[0010] Remote monitoring and intervention module: It is used for medical staff to view the patient's exercise tolerance data in real time through the system interface or mobile application. When the patient's exercise tolerance data is abnormal, the system will automatically trigger the early warning mechanism to remind medical staff to pay attention to the patient's condition and intervene.
[0011] Furthermore, the data cleaning uses a standard deviation method to identify and process outliers.
[0012] Furthermore, the data normalization adopts improved Z-score normalization.
[0013] Furthermore, the data normalization uses minimum-maximum normalization.
[0014] Furthermore, the exercise tolerance data is normalized.
[0015] Furthermore, the WD-RCNN model evaluation process is as follows: first, the input exercise tolerance data is subjected to wavelet transform to extract the time-frequency features of the exercise tolerance data, and then the features are learned and the model is trained through a convolutional neural network. The model outputs the patient's exercise tolerance level score, which is divided into three categories: low, medium, and high.
[0016] Furthermore, the time-frequency characteristics of the exercise tolerance data include extracting key features from physiological parameters and exercise data, including heart rate change rate, blood pressure range, blood oxygen saturation change, maximum acceleration, and exercise pattern distribution.
[0017] Furthermore, the fitting and prediction effects of the WD-RCNN model were evaluated by the coefficient of determination (R 2 ), root mean square error (RMSE) and relative root mean square error (rRMSE) were used as evaluation indicators.
[0018] Furthermore, the moving average method is used for data sequences.
[0019] Furthermore, the moving average method calculation steps are specifically as follows:
[0020] (1) Determine the time window size n of the moving average; the window size is determined according to specific application requirements;
[0021] (2) Starting from the first data point in the data sequence, the moving average is calculated sequentially. For the first moving average, the first n data points need to be used for calculation; for subsequent moving averages, slide back one data point each time and recalculate the average.
[0022] Compared with the prior art, the present invention has the following beneficial effects:
[0023] The present invention sets reasonable thresholds and cleaning algorithms to remove outliers, ensure the accuracy and reliability of data, and standardize data from different sources and units so that they can be compared and analyzed on the same scale. The exercise tolerance assessment algorithm can accurately assess the patient's exercise tolerance level and predict its changing trend, which helps medical staff to promptly discover possible health risks of patients and take corresponding measures for prevention and treatment. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] Figure 1 This is a structural diagram of the atrial fibrillation patient health management system of the present invention. DETAILED DESCRIPTION
[0025] The present invention is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several changes and improvements can also be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.
[0026] like Figure 1 As shown in the figure, the health management system for patients with atrial fibrillation under the auxiliary feedback of exercise tolerance parameters includes:
[0027] Exercise tolerance monitoring module: used to collect the patient's physiological parameters in real time through wearable sensors and obtain the patient's exercise parameters using accelerometers and gyroscope sensors. Physiological parameters include the patient's heart rate H, blood pressure P, and blood oxygen saturation β; exercise parameters include exercise intensity ν and exercise mode M. Physiological parameters and exercise parameters are collectively referred to as exercise tolerance data;
[0028] Data processing module: used for data cleaning, data standardization and data normalization of exercise tolerance data;
[0029] Data cleaning: By setting reasonable thresholds and algorithms, outliers are removed to ensure data accuracy and reliability. The standard deviation method is used to identify and process outliers. For each data point x i (Exercise tolerance data x i , i∈{1,2,3...,N}), calculate its difference with the average value The difference χi =x i -u, and with standard deviation If the difference χ i If the value of -s exceeds the threshold m=3s (3 times the standard deviation), the data point is considered an outlier and is removed or replaced.
[0030] Data standardization: Standardize data from different sources and units so that they can be compared and analyzed on the same scale.
[0031] Data normalization method: Improved Z-score normalization.
[0032]
[0033] Among them, x is the exercise tolerance data after data cleaning, μ is the mean of the data, and σ is the standard deviation of the data. After standardization, the mean of the data is 0 and the standard deviation is 1.
[0034] Data normalization: Through the normalization algorithm, the data is converted into a dimensionless value between 0 and 1, which is convenient for subsequent algorithm processing and analysis.
[0035] Minimum-maximum normalization formula:
[0036] Among them, x is the exercise tolerance data after data cleaning and standardization, x min and x max are the minimum and maximum values in the data respectively. After normalization, the data value range is between 0 and 1.
[0037] Exercise tolerance datax i =λ H ·H i +λ P ·P i +λ β β i +λ ν ·ν i +λ M ·M i , where λ is the normalization coefficient of the corresponding parameter.
[0038] Data analysis module: perform WD-RCNN model evaluation on the exercise tolerance data, and predict the change trend of the patient's exercise tolerance level in the future by analyzing the patient's historical exercise tolerance data and using the moving average method;
[0039] Extract key features from physiological parameters: heart rate change rate, blood pressure range, and blood oxygen saturation changes.
[0040] Extract key features from motion data: average maximum acceleration, motion pattern distribution.
[0041] Exercise tolerance assessment model: A machine learning based model WD-RCNN was established using the extracted features as input.
[0042] The WD-RCNN model is a convolutional neural network model based on wavelet decomposition. The model first performs wavelet transform on the input data to extract the time-frequency features of the data, and then learns the features and trains the model through the convolutional neural network. The WD-RCNN model combines the multi-scale feature extraction capability of wavelet decomposition and the deep learning capability of convolutional neural networks. By introducing wavelet decomposition and multi-scale convolution, the WD-RCNN model can better represent complex data and improve the performance and generalization ability of the model.
[0043] The specific network structure is as follows:
[0044]
[0045]
[0046] After calculating various data features as above, training begins, using the WD-RCNN neural network model; it includes: convolution layer 1, pooling layer 1, convolution layer 2, pooling layer 2, ... convolution layer 6, pooling layer 6, fully connected layer and output layer. Data annotation uses 70% of the total amount of each feature data as a training set and 30% as a test set. The output layer is divided into two categories: faulty category and non-faulty category.
[0047] The model fit and prediction performance were estimated by using the coefficient of determination (R 2 ), root mean square error (RMSE) and relative root mean square error (rRMSE) were used as evaluation indicators.
[0048]
[0049]
[0050] Where, L i The exercise tolerance data is in the normal range; The average value of exercise tolerance data in the normal range; is the model estimate; n is the sample size.
[0051] Model input: exercise tolerance data, train the model to identify patterns related to exercise tolerance and predict exercise tolerance level. Model output: The model outputs the patient's exercise tolerance level score, which is divided into three categories: low, medium, and high.
[0052] Moving average method is a commonly used statistical analysis method for data series (exercise tolerance data x i , i∈{1,2,3...,N})) for smoothing and trend prediction. The calculation formula is: moving average = (data point 1 + data point 2 + data point 3 +... + data point n) / n
[0053] Among them, n is the time window size of the moving average, which means taking the first n data points for average calculation. The main function of the moving average method is to reduce the random fluctuation of data, make the trend more obvious, and facilitate analysis and prediction.
[0054] The calculation steps are as follows: Determine the time window size n of the moving average. This window size is determined according to the specific application requirements. Generally, it needs to be selected according to the periodicity and volatility of the data. Starting from the first data point of the data sequence, the moving average is calculated in sequence. For the first moving average, the first n data points need to be used for calculation; for subsequent moving averages, slide back one data point each time and recalculate the average.
[0055] Feedback and health management module: formulate appropriate exercise plans according to the patient's exercise tolerance level, and push personalized health management plans and precautions to patients and medical staff in real time through mobile applications or smart devices;
[0056] Remote monitoring and intervention module: It is used for medical staff to view the patient's exercise tolerance data in real time through the system interface or mobile application. When the patient's exercise tolerance data is abnormal, the system will automatically trigger the early warning mechanism to remind medical staff to pay attention to the patient's condition and intervene.
[0057] The present invention collects and analyzes the patient's exercise tolerance parameters in real time through wearable sensors, so that the health management system can grasp the patient's health status in real time and provide accurate health management and rehabilitation guidance. The system can formulate personalized health management and rehabilitation plans according to the patient's exercise tolerance level, and use artificial intelligence to dynamically adjust according to the patient's actual situation. This helps to improve the pertinence and effectiveness of treatment plans, supports medical staff to remotely monitor the patient's health status, and conduct remote intervention when necessary. This helps to improve the utilization efficiency of medical resources and reduce medical costs.
[0058] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essence of the present invention. In the absence of conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.
Claims
1. A health management system for patients with atrial fibrillation under the auxiliary feedback of exercise tolerance parameters, characterized in that: include: Exercise tolerance monitoring module: used to collect the patient's physiological parameters in real time through wearable sensors and obtain the patient's exercise parameters using accelerometers and gyroscope sensors. The physiological parameters include the patient's heart rate H, blood pressure P, and blood oxygen saturation β; the exercise parameters include exercise intensity ν and exercise mode M. Physiological parameters and exercise parameters are collectively referred to as exercise tolerance data; Data processing module: used for data cleaning, data standardization and data normalization of the exercise tolerance data; Data analysis module: perform WD-RCNN model evaluation on the exercise tolerance data, and predict the change trend of the patient's exercise tolerance level in the future by analyzing the patient's historical exercise tolerance data and using the moving average method; Feedback and health management module: formulate appropriate exercise plans according to the patient's exercise tolerance level, and push personalized health management plans and precautions to patients and medical staff in real time through mobile applications or smart devices; Remote monitoring and intervention module: It is used for medical staff to view the patient's exercise tolerance data in real time through the system interface or mobile application. When the patient's exercise tolerance data is abnormal, the system will automatically trigger the early warning mechanism to remind medical staff to pay attention to the patient's condition and intervene.
2. The atrial fibrillation patient health management system with exercise tolerance parameter auxiliary feedback according to claim 1, characterized in that: The data cleaning uses the standard deviation method to identify and process outliers.
3. The atrial fibrillation patient health management system with exercise tolerance parameter auxiliary feedback according to claim 1, characterized in that: The data normalization adopts improved Z-score normalization.
4. The atrial fibrillation patient health management system with exercise tolerance parameter auxiliary feedback according to claim 1, characterized in that: The data normalization uses min-max normalization.
5. The health management system for patients with atrial fibrillation under the auxiliary feedback of exercise tolerance parameters according to claim 1, characterized in that: The exercise tolerance data after data normalization.
6. The health management system for patients with atrial fibrillation under the auxiliary feedback of exercise tolerance parameters according to claim 1, characterized in that: The evaluation process of the WD-RCNN model is as follows: first, the input exercise tolerance data is subjected to wavelet transform to extract the time-frequency features of the exercise tolerance data, and then the features are learned and the model is trained through a convolutional neural network. The model outputs the patient's exercise tolerance level score, which is divided into three categories: low, medium, and high.
7. The atrial fibrillation patient health management system with exercise tolerance parameter auxiliary feedback according to claim 6, characterized in that: The time-frequency characteristics of the exercise tolerance data include extracting key features from physiological parameters and exercise data, including heart rate change rate, blood pressure range, blood oxygen saturation change, maximum acceleration, and exercise pattern distribution.
8. The atrial fibrillation patient health management system with exercise tolerance parameter auxiliary feedback according to claim 6, characterized in that: The fitting and prediction performance of the WD-RCNN model were measured by the coefficient of determination (R 2 ), root mean square error (RMSE) and relative root mean square error (rRMSE) were used as evaluation indicators.
9. Implementing the atrial fibrillation patient health management system under the exercise tolerance parameter auxiliary feedback of claim 1, characterized in that: The moving average method is applied to the data series.
10. The atrial fibrillation patient health management system with exercise tolerance parameter auxiliary feedback according to claim 9, characterized in that: The moving average calculation steps are as follows: (1) Determine the time window size n of the moving average; the window size is determined according to specific application requirements; (2) Starting from the first data point in the data sequence, the moving average is calculated sequentially. For the first moving average, the first n data points need to be used for calculation; for subsequent moving averages, slide back one data point each time and recalculate the average.
