Cardiovascular disease health management system based on user behaviors

Abstract

The invention relates to the technical field of medical health management, in particular to a cardiovascular disease health management system based on user behaviors. Comprising a user side, a multi-modal data acquisition layer, a data fusion module, a preprocessing module, an encryption transmission unit, a processor, a disease prediction module, an intervention scheme generation module, a multi-channel push module and an intervention execution unit. The data fusion module is connected with the multi-modal data acquisition layer, the preprocessing module is connected with the data fusion module, the encryption transmission unit is connected with the preprocessing module and the processor, the disease prediction module is connected with the processor, and the intervention scheme generation module is connected with the disease prediction module. In this way, the technical problems that a cardiovascular disease management system adopted in the prior art only depends on single physiological data prediction and lacks behavior and environment fusion analysis, and after disease prediction, the intervention means is single are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical and health management, and particularly to a cardiovascular disease health management system based on user behavior. Background Art

[0002] Cardiovascular diseases are one of the major health threats globally, and their onset is often closely related to patients' life behaviors, environmental factors, etc. Traditional cardiovascular disease health management models mainly focus on treatment and rehabilitation after the occurrence of diseases, and pay insufficient attention to prevention and risk intervention in the early stage of diseases.

[0003] For example, a cardiovascular disease prediction system disclosed in the prior art patent application number CN202110617968.9 and a cardiovascular disease management system including the prediction system enable residents without diseases to predict the probability of chronic cardiovascular diseases; residents with diseases can predict the probability of downstream cardiovascular diseases, communicate with medical staff, receive corresponding further evaluations and medical suggestions, receive chronic disease prevention and control knowledge, and conduct patient education.

[0004] However, the cardiovascular disease management system adopted in the above method only relies on single physiological data prediction, lacks the integration analysis of behavior and environment, and has a single intervention means after disease prediction. Summary of the Invention

[0005] The purpose of the present invention is to provide a cardiovascular disease health management system based on user behavior, aiming to solve the technical problems that the cardiovascular disease management system adopted in the prior art only relies on single physiological data prediction, lacks the integration analysis of behavior and environment, and has a single intervention means after disease prediction.

[0006] To achieve the above purpose, a cardiovascular disease health management system based on user behavior adopted by the present invention includes a user terminal, a multi-modal data acquisition layer, a data fusion module, a preprocessing module, an encrypted transmission unit, a processor, a disease prediction module, an intervention plan generation module, a multi-channel push module, and an intervention execution unit; the data fusion module is connected to the multi-modal data acquisition layer, the preprocessing module is connected to the data fusion module, the encrypted transmission unit is connected to both the preprocessing module and the processor, the disease prediction module is connected to the processor, the intervention plan generation module is connected to the disease prediction module, both the multi-channel push module and the intervention execution unit are connected to the intervention plan generation module, and the user terminal is connected to the multi-channel push module;

[0007] The multi-modal data acquisition layer is used to comprehensively collect multi-dimensional data of users;

[0008] The data fusion module is used to fuse the collected multi-dimensional user data and transmit it to the preprocessing module;

[0009] The preprocessing module is used to clean the fused heterogeneous data and remove noise data and outliers;

[0010] The encrypted transmission unit is used to encrypt and transmit the preprocessed data to the processor;

[0011] The processor uses the disease prediction module to accurately predict the risk of cardiovascular disease occurrence in combination with the received data;

[0012] The intervention plan generation module generates a personalized health intervention plan according to the risk prediction results and user individual characteristics by means of a reinforcement learning algorithm;

[0013] The multi-channel push module is used to push the intervention plan to the user terminal, and the intervention execution unit facilitates the user to make actual action interventions.

[0014] Among them, the multi-modal data acquisition layer includes a behavior data acquisition module, an environmental perception module, a physiological data acquisition module, and a psychological data acquisition module. The behavior data acquisition module, the environmental perception module, the physiological data acquisition module, and the psychological data acquisition module are all connected to the data fusion module;

[0015] The behavior data acquisition module, the environmental perception module, the physiological data acquisition module, and the psychological data acquisition module work together to comprehensively collect the user's daily behaviors, physiological indicators, psychological states, and environmental data.

[0016] Among them, the encrypted transmission unit includes an encryption module, a decryption module, a transmission module, a key management module, and a key generation module. The encryption module and the decryption module are respectively connected to the preprocessing module and the processor. The transmission module is connected to both the preprocessing module and the processor. The key management module is connected to the encryption module, the decryption module, and the key generation module.

[0017] Among them, the intervention execution unit includes a VR behavior correction module and a medication recommendation generation module. The VR behavior correction module and the medication recommendation generation module are both connected to the intervention plan generation module.

[0018] Among them, the cardiovascular disease health management system based on user behavior further includes a face acquisition module and an emotion analysis module. The face acquisition module is connected to the data fusion module through the emotion analysis module.

[0019] Among them, the cardiovascular disease health management system based on user behavior further includes a feedback module and an acquisition module. The feedback module is connected to both the user terminal and the processor, and the acquisition module is connected to the processor;

[0020] The feedback module is used to collect the feedback on the implementation of the intervention plan by the user and the subjective feelings, conduct sentiment analysis and intention recognition on the feedback information through natural language processing technology, evaluate the effectiveness of the intervention plan and the user satisfaction, and push them to the processor to perform online update and optimization on the intervention plan generation algorithm;

[0021] The acquisition module is used to regularly update the historical data, so that the generation of the intervention plan can be adaptively optimized.

[0022] Among them, the cardiovascular disease health management system based on user behavior further includes a storage module, and the storage module is connected to the processor;

[0023] The storage module uses HDFS (Hadoop Distributed File System) to store the data in the whole system. The distributed file system can be used to store a large amount of image data and log files.

[0024] Among them, the cardiovascular disease health management system based on user behavior further includes a compression module, and the compression module is connected to the storage module;

[0025] The compression module uses a lossless compression algorithm to compress the structured data of the user;

[0026] For the unstructured or semi-structured data of the user, a lossless or lossy compression algorithm is selected based on the importance and accuracy requirements of the data.

[0027] A cardiovascular disease health management system based on user behavior according to the present invention. First, the present invention comprehensively collects multi-dimensional data of users through the multi-modal data acquisition layer, and uses the data fusion module to fuse the collected multi-dimensional data of users, and then transmits it to the preprocessing module. The preprocessing module is used to clean the fused heterogeneous data, remove noise data and outliers, and then the encryption transmission unit is used to encrypt and transmit the preprocessed data to the processor; the processor uses the disease prediction module to accurately predict the occurrence risk of cardiovascular diseases in combination with the received data; the intervention plan generation module generates a personalized health intervention plan by means of a reinforcement learning algorithm according to the risk prediction result and user individual characteristics; the multi-channel push module is used to push the intervention plan to the user terminal, and the intervention execution unit facilitates the user to make actual action interventions. In this way, the technical problems in the prior art that the cardiovascular disease management system only relies on single physiological data prediction, lacks behavioral and environmental fusion analysis, and has a single intervention means after disease prediction are solved.

[0028] The present invention comprehensively collects multi-dimensional data of users through the multi-modal data acquisition layer, including physiological data, behavioral data, environmental data, etc. Using the data fusion module to perform fusion processing on these data can more comprehensively reflect the health status and potential risks of users. By combining multi-dimensional data, the disease prediction module of the present invention can more accurately predict the occurrence risk of cardiovascular diseases and improve the accuracy of prediction.

[0029] After disease prediction in the present invention, the intervention plan generation module will generate a personalized health intervention plan by means of a reinforcement learning algorithm according to the risk prediction result and user individual characteristics (such as age, gender, health status, etc.). This personalized intervention plan can more accurately meet the needs of users and improve the pertinence and effectiveness of the intervention.

[0030] The present invention uses the encryption transmission unit to encrypt and transmit the preprocessed data to ensure the security and privacy protection of the data during the transmission process. This helps to enhance the user's trust in the system and improve the reliability and usability of the system.

[0031] The intervention execution unit of the present invention facilitates the user to make actual action interventions. By providing specific operation steps and a feedback mechanism, the intervention execution unit can prompt the user to transform the intervention plan into actual actions, thereby more effectively reducing the occurrence risk of cardiovascular diseases. Brief Description of the Drawings

[0032] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0033] Figure 1 is the principle block diagram of the first embodiment of the present invention.

[0034] Figure 2 is the principle block diagram of the second embodiment of the present invention.

[0035] Figure 3 is the principle block diagram of the third embodiment of the present invention.

[0036] 101 - Client, 102 - Multimodal Data Acquisition Layer, 103 - Data Fusion Module, 104 - Preprocessing Module, 105 - Encrypted Transmission Unit, 106 - Processor, 107 - Disease Prediction Module, 108 - Intervention Plan Generation Module, 109 - Multi-channel Push Module, 110 - Intervention Execution Unit, 111 - Behavior Data Acquisition Module, 112 - Environment Sensing Module, 113 - Physiological Data Acquisition Module, 114 - Psychological Data Acquisition Module, 115 - Encryption Module, 116 - Decryption Module, 117 - Transmission Module, 118 - Key Management Module, 119 - Key Generation Module, 120 - VR Behavior Correction Module, 121 - Dispensing Recommendation Generation Module, 201 - Facial Acquisition Module, 202 - Emotion Analysis Module, 203 - Feedback Module, 204 - Acquisition Module, 205 - Storage Module, 206 - Compression Module, 301 - Tamper-proof Module, 302 - Management Module, 303 - Login Module, 304 - Permission Assignment Module Detailed Embodiment

[0037] The following will describe in detail the embodiments of the present invention. The examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to explain the present invention, and should not be construed as a limitation to the present invention.

[0038] The first embodiment of this application is as follows:

[0039] Please refer to Figure 1 , Figure 1 is the principle block diagram of the first embodiment of the present invention.

[0040] The present invention provides a cardiovascular disease health management system based on user behavior, including a user terminal 101, a multimodal data acquisition layer 102, a data fusion module 103, a preprocessing module 104, an encryption transmission unit 105, a processor 106, a disease prediction module 107, an intervention plan generation module 108, a multi-channel push module 109, and an intervention execution unit 110; the multimodal data acquisition layer 102 includes a behavior data acquisition module 111, an environmental perception module 112, a physiological data acquisition module 113, and a psychological data acquisition module 114, the encryption transmission unit 105 includes an encryption module 115, a decryption module 116, a transmission module 117, a key management module 118, and a key generation module 119, the intervention execution unit 110 includes a VR behavior correction module 120 and a dispensing advice generation module 121. The foregoing solution solves the technical problems in the prior art that the cardiovascular disease management system only relies on single physiological data prediction, lacks the fusion analysis of behavior and environment, and has a single intervention means after disease prediction.

[0041] For this specific embodiment, the multimodal data acquisition layer 102 is used to comprehensively collect multi-dimensional data of the user;

[0042] The data fusion module 103 is used to fuse the collected multi-dimensional data of the user and transmit it to the preprocessing module 104;

[0043] The preprocessing module 104 is used to clean the fused heterogeneous data and remove noise data and outliers;

[0044] The encryption transmission unit 105 is used to encrypt and transmit the preprocessed data to the processor 106;

[0045] The processor 106 uses the disease prediction module 107 to accurately predict the risk of cardiovascular disease occurrence in combination with the received data;

[0046] The intervention plan generation module 108 generates a personalized health intervention plan according to the risk prediction result and the user's individual characteristics by means of a reinforcement learning algorithm;

[0047] The multi-channel push module 109 is used to push the intervention plan to the user terminal 101, and the intervention execution unit 110 facilitates the user to make actual action interventions.

[0048] Among them, the data fusion module 103 is connected to the multi-modal data acquisition layer 102, the preprocessing module 104 is connected to the data fusion module 103, the encryption transmission unit 105 is connected to both the preprocessing module 104 and the processor 106, the disease prediction module 107 is connected to the processor 106, the intervention plan generation module 108 is connected to the disease prediction module 107, both the multi-channel push module 109 and the intervention execution unit 110 are connected to the intervention plan generation module 108, and the user terminal 101 is connected to the multi-channel push module 109. First, the present invention comprehensively collects multi-dimensional data of users through the multi-modal data acquisition layer 102, and uses the data fusion module 103 to fuse the collected multi-dimensional data of users, and then transmits it to the preprocessing module 104. The preprocessing module 104 is used to clean the fused heterogeneous data, remove noise data and outliers. Then, the encryption transmission unit 105 is used to encrypt and transmit the preprocessed data to the processor 106; the processor 106 uses the disease prediction module 107 to accurately predict the risk of cardiovascular disease occurrence by combining the received data; the intervention plan generation module 108 generates a personalized health intervention plan according to the risk prediction result and user individual characteristics with the help of a reinforcement learning algorithm; the multi-channel push module 109 is used to push the intervention plan to the user terminal 101, and the intervention execution unit 110 facilitates the user to make actual action interventions, thus solving the technical problems in the prior art that the cardiovascular disease management system only relies on single physiological data prediction, lacks behavioral and environmental fusion analysis, and has a single intervention means after disease prediction;

[0049] The disease prediction module 107 constructs a deep neural network model based on large-scale historical case data and current user data, comprehensively considers the complex associations among user behavior, physiology, psychology, and environmental factors, and realizes accurate prediction of the risk of cardiovascular disease occurrence. The model uses a convolutional neural network (CNN) to extract features from user behavior data and physiological data, and uses a long short-term memory network (LSTM) to model time series data to capture the dynamic change trend of user data, and combines an attention mechanism to highlight the influence of key features on the prediction result, thereby improving the accuracy and robustness of the prediction.

[0050] Secondly, the multi-modal data acquisition layer 102 includes a behavior data acquisition module 111, an environment perception module 112, a physiological data acquisition module 113, and a psychological data acquisition module 114. The behavior data acquisition module 111, the environment perception module 112, the physiological data acquisition module 113, and the psychological data acquisition module 114 are all connected to the data fusion module 103;

[0051] The behavior data collection module 111, the environmental perception module 112, the physiological data collection module 113, and the psychological data collection module 114 work together to comprehensively collect the user's daily behaviors, physiological indicators, psychological states, and environmental data. For example, the behavior data collection module 111 collects information such as the user's exercise steps, sleep duration and quality through a smart bracelet, and records the user's eating habits and smoking and drinking situations through a mobile application; the physiological data collection module 113 uses devices such as a smart sphygmomanometer and a blood glucose meter to collect physiological data such as the user's heart rate, blood pressure, and blood glucose; the psychological data collection module 114 obtains the user's psychological stress and emotional state data by means of psychological assessment questionnaires and voice emotion analysis techniques; the environmental perception module 112 monitors environmental factor data such as the temperature, humidity, and PM2.5 concentration of the environment where the user is located in real time.

[0052] At the same time, the encryption module 115 and the decryption module 116 are respectively connected to the preprocessing module 104 and the processor 106, the transmission module 117 is connected to both the preprocessing module 104 and the processor 106, and the key management module 118 is connected to the encryption module 115, the decryption module 116, and the key generation module 119. When specifically used, first, the key generation module 119 generates a key pair for encryption and decryption through a specific key generation algorithm, such as a random number generator combined with a specific hash function. These keys are passed to the key management module 118 for secure storage and management. Then, the encryption module 115 adopts a symmetric or asymmetric encryption algorithm (such as AES, RSA, etc.), and uses the encryption key provided by the key management module 118 to encrypt the preprocessed data. The encryption process includes multiple rounds of substitution, permutation, and mixing operations to ensure the confidentiality and anti-attack ability of the data.

[0053] The encrypted data is transmitted to the processor 106 through the transmission module 117, and the transmission module 117 adopts a secure transmission protocol (such as TLS) to further protect the security of the data during transmission. At the receiving end, the processor 106 receives the encrypted data and passes it to the decryption module 116.

[0054] The decryption module 116 uses the decryption key provided by the key management module 118 and adopts a decryption algorithm corresponding to the encryption module 115 to decrypt the encrypted data. The decryption process needs to strictly follow the inverse process of the encryption algorithm to ensure that the original data can be correctly restored. The processor 106 obtains the decrypted data and performs further processing or application.

[0055] In addition, the intervention execution unit 110 includes a VR behavior correction module 120 and a dispensing advice generation module 121, and both the VR behavior correction module 120 and the dispensing advice generation module 121 are connected to the intervention plan generation module 108;

[0056] Based on the plan generated by the intervention plan generation module 108, the VR behavior correction module 120 simulates high-risk scenarios for the user to perform immersive training;

[0057] Similarly based on the plan generated by the intervention plan generation module 108, the dispensing advice generation module 121 realizes the precise preparation of drugs and generates dispensing advice in an automated and intelligent manner.

[0058] When using a cardiovascular disease health management system based on user behavior in this embodiment, the present invention first comprehensively collects multi-dimensional data of the user through the multi-modal data acquisition layer 102, and uses the data fusion module 103 to fuse the collected multi-dimensional data of the user, and then transmits it to the preprocessing module 104. The preprocessing module 104 is used to clean the fused heterogeneous data, remove noise data and outliers, and then the encryption transmission unit 105 is used to encrypt and transmit the preprocessed data to the processor 106; the processor 106 uses the disease prediction module 107 to accurately predict the occurrence risk of cardiovascular disease in combination with the received data; the intervention plan generation module 108 generates a personalized health intervention plan according to the risk prediction result and the user's individual characteristics by means of a reinforcement learning algorithm; the multi-channel push module 109 is used to push the intervention plan to the user terminal 101, and the intervention execution unit 110 facilitates the user to make actual action interventions, thereby solving the technical problems in the prior art that the cardiovascular disease management system only relies on single physiological data prediction, lacks the fusion analysis of behavior and environment, and has a single intervention means after disease prediction.

[0059] The present invention comprehensively collects multi-dimensional data of the user through the multi-modal data acquisition layer 102, including physiological data, behavior data, environmental data, etc. Using the data fusion module 103 to perform fusion processing on these data can more comprehensively reflect the user's health status and potential risks. By combining multi-dimensional data, the disease prediction module 107 of the present invention can more accurately predict the occurrence risk of cardiovascular disease and improve the accuracy of prediction.

[0060] After disease prediction, the intervention plan generation module 108 of the present invention will generate a personalized health intervention plan according to the risk prediction result and the user's individual characteristics (such as age, gender, health status, etc.) by means of a reinforcement learning algorithm. This personalized intervention plan can more accurately meet the needs of the user and improve the pertinence and effectiveness of the intervention.

[0061] The present invention uses the encryption and transmission unit 105 to encrypt and transmit the preprocessed data, ensuring the security and privacy protection of the data during the transmission process. This helps to enhance the user's trust in the system and improve the reliability and availability of the system.

[0062] The intervention execution unit 110 of the present invention facilitates the user to make actual action interventions. By providing specific operation steps and a feedback mechanism, the intervention execution unit 110 can prompt the user to transform the intervention plan into actual actions, thereby more effectively reducing the risk of cardiovascular diseases.

[0063] The second embodiment of the present application is:

[0064] Based on the first embodiment, please refer to Figure 2 , Figure 2 which is the principle block diagram of the second embodiment of the present invention.

[0065] The present invention provides a cardiovascular disease health management system based on user behavior, further including a face acquisition module 201, an emotion analysis module 202, a feedback module 203, an acquisition module 204, a storage module 205, and a compression module 206.

[0066] For this specific embodiment, the face acquisition module 201 is connected to the data fusion module 103 through the emotion analysis module 202. The face acquisition module 201 uses an algorithm based on computer vision to capture the user's facial expressions. It tracks the facial feature points of the user in real time, and by comparing the changes in facial feature points in different emotional states, it establishes a mapping relationship between facial expressions and emotional states;

[0067] The emotion analysis module 202 uses machine learning or deep learning algorithms to analyze and process the collected facial expression data, automatically identify the emotional features in the facial expressions, such as frowning indicating anger or dissatisfaction, smiling indicating happiness or satisfaction, etc., and combines NLP technology to analyze social media data to evaluate the impact of psychological stress on cardiovascular health.

[0068] Among them, the feedback module 203 is connected to both the user terminal 101 and the processor 106, and the acquisition module 204 is connected to the processor 106;

[0069] The feedback module 203 is used to collect the feedback on the execution of the intervention plan by the user and the subjective feelings, perform sentiment analysis and intention recognition on the feedback information through natural language processing technology, evaluate the effectiveness of the intervention plan and the user satisfaction, and push it to the processor 106 to perform online update and optimization on the intervention plan generation algorithm;

[0070] The acquisition module 204 is used to update historical data regularly, so that the generation of the intervention plan can be optimized adaptively.

[0071] Secondly, the storage module 205 is connected to the processor 106;

[0072] The storage module 205 stores data in the entire system using HDFS (Hadoop Distributed File System). The distributed file system can be used to store a large amount of image data and log files.

[0073] The compression module 206 is connected to the storage module 205;

[0074] The compression module 206 compresses the structured data of users using a lossless compression algorithm;

[0075] For the unstructured or semi-structured data of users, a lossless or lossy compression algorithm is selected based on the importance and accuracy requirements of the data.

[0076] Using a cardiovascular disease health management system based on user behavior in this embodiment, the face acquisition module 201 uses an algorithm based on computer vision to capture the facial expressions of users. Real-time tracking of the facial feature points of users, and by comparing the changes in facial feature points in different emotional states, a mapping relationship between facial expressions and emotional states is established;

[0077] The emotion analysis module 202 uses machine learning or deep learning algorithms to analyze and process the collected facial expression data, automatically identify the emotion features in the facial expressions, such as frowning indicating anger or dissatisfaction, smiling indicating happiness or satisfaction, etc., and combines NLP technology to analyze social media data to evaluate the impact of psychological stress on cardiovascular health.

[0078] The third embodiment of this application is:

[0079] On the basis of the second embodiment, please refer to Figure 3 , Figure 3 is the principle block diagram of the third embodiment of the present invention.

[0080] The present invention provides a cardiovascular disease health management system based on user behavior, further including an anti-tampering module 301, a maintenance module 302, a login module 303, and a permission allocation module 304.

[0081] For this specific embodiment, the anti-tampering module 301 is connected to the storage module 205, and the anti-tampering module 301 is used to prevent the data in the storage module 205 from being tampered with.

[0082] Among them, the maintenance module 302 is connected to the processor 106, the login module 303 is connected to the maintenance module 302, the permission allocation module 304 is connected to the login module 303. The maintenance module 302 is used for managers to manage the processor 106 and the entire system. The login module 303 is used for managers to log in to the maintenance module 302. The permission allocation module 304 is used to allocate different management permissions according to the identities of the logged-in personnel.

[0083] Using the cardiovascular disease health management system based on user behavior in this embodiment, the anti-tampering module 301 is used to prevent the data in the storage module 205 from being tampered with. The maintenance module 302 is used for managers to manage the processor 106 and the entire system. The login module 303 is used for managers to log in to the maintenance module 302. The permission allocation module 304 is used to allocate different management permissions according to the identities of the logged-in personnel.

[0084] The above-disclosed is only a preferred embodiment of the present invention. Of course, the scope of the rights of the present invention cannot be limited thereby. Those of ordinary skill in the art can understand all or part of the processes of implementing the above embodiments, and the equivalent changes made according to the claims of the present invention still fall within the scope covered by the present invention.

Claims

1. A cardiovascular disease health management system based on user behavior, characterized in that it includes a user terminal, a multi-modal data acquisition layer, a data fusion module, a preprocessing module, an encrypted transmission unit, a processor, a disease prediction module, an intervention plan generation module, a multi-channel push module, and an intervention execution unit; the data fusion module is connected to the multi-modal data acquisition layer, the preprocessing module is connected to the data fusion module, the encrypted transmission unit is connected to both the preprocessing module and the processor, the disease prediction module is connected to the processor, the intervention plan generation module is connected to the disease prediction module, the multi-channel push module and the intervention execution unit are both connected to the intervention plan generation module, and the user terminal is connected to the multi-channel push module; The multi-modal data acquisition layer is used to comprehensively collect multi-dimensional data of users; The data fusion module is used to fuse the collected multi-dimensional data of users and transmit it to the preprocessing module; The preprocessing module is used to clean the fused heterogeneous data and remove noise data and outliers; The encrypted transmission unit is used to encrypt and transmit the preprocessed data to the processor; The processor uses the disease prediction module to accurately predict the risk of cardiovascular disease occurrence in combination with the received data; The intervention plan generation module generates a personalized health intervention plan according to the risk prediction result and user individual characteristics with the help of a reinforcement learning algorithm; The multi-channel push module is used to push the intervention plan to the user terminal, and the intervention execution unit facilitates users to make actual action interventions.

2. The cardiovascular disease health management system based on user behavior according to claim 1, characterized in that the multi-modal data acquisition layer includes a behavior data acquisition module, an environmental perception module, a physiological data acquisition module, and a psychological data acquisition module, and the behavior data acquisition module, the environmental perception module, the physiological data acquisition module, and the psychological data acquisition module are all connected to the data fusion module; The behavior data acquisition module, the environmental perception module, the physiological data acquisition module, and the psychological data acquisition module work together to comprehensively collect users' daily behaviors, physiological indicators, psychological states, and environmental data.

3. The cardiovascular disease health management system based on user behavior according to claim 2, characterized in that the encrypted transmission unit includes an encryption module, a decryption module, a transmission module, a key management module, and a key generation module, the encryption module and the decryption module are respectively connected to the preprocessing module and the processor, the transmission module is connected to both the preprocessing module and the processor, and the key management module is connected to the encryption module, the decryption module, and the key generation module.

4. The cardiovascular disease health management system based on user behavior according to claim 3, characterized in that the intervention execution unit includes a VR behavior correction module and a medication advice generation module, and the VR behavior correction module and the medication advice generation module are both connected to the intervention plan generation module.

5. The cardiovascular disease health management system based on user behavior according to claim 4, wherein the cardiovascular disease health management system based on user behavior further includes a face acquisition module and an emotion analysis module, and the face acquisition module is connected to the data fusion module through the emotion analysis module.

6. The cardiovascular disease health management system based on user behavior according to claim 5, wherein the cardiovascular disease health management system based on user behavior further includes a feedback module and an acquisition module. The feedback module is connected to both the user terminal and the processor, and the acquisition module is connected to the processor; the feedback module is used to collect the feedback on the implementation of the intervention plan by the user and the subjective feelings, perform sentiment analysis and intention recognition on the feedback information through natural language processing technology, evaluate the effectiveness of the intervention plan and the user satisfaction, and push it to the processor to perform online update and optimization on the intervention plan generation algorithm; the acquisition module is used to regularly update the historical data so that the generation of the intervention plan can be adaptively optimized.

7. The cardiovascular disease health management system based on user behavior according to claim 6, wherein the cardiovascular disease health management system based on user behavior further includes a storage module, and the storage module is connected to the processor; the storage module uses HDFS (Hadoop Distributed File System) to store the data in the whole system. The distributed file system can be used to store a large amount of image data and log files.

8. The cardiovascular disease health management system based on user behavior according to claim 7, wherein the cardiovascular disease health management system based on user behavior further includes a compression module, and the compression module is connected to the storage module; the compression module uses a lossless compression algorithm to compress the structured data of the user; for the unstructured or semi-structured data of the user, a lossless or lossy compression algorithm is selected based on the importance and accuracy requirements of the data.

Images