Comprehensive health management modeling system and method based on visual management

Abstract

The invention discloses a comprehensive health management modeling system and method based on visual management, and belongs to the technical field of health management systems.The method specifically comprises the steps that physiological, activity and diet data of a user are collected through an automatic means, health risks are recognized by means of a survival analysis strategy after preprocessing, and a comprehensive health management model is obtained by combining machine learning and a statistical method; a risk assessment model is constructed, whether data is abnormal or not is judged in real time, for abnormal data, a natural language processing technology is used for feeding back to a user, for health data, a future health state is predicted through causal reasoning, then a personalized health management scheme is formulated according to user data, gene information and environmental factors, and the effect of the scheme is assessed through noninvasive monitoring. And dynamic adjustment is performed according to the evaluation result, so that the real-time performance and effectiveness of health management are ensured, the actual requirements of the user can be better met, and the health management effect is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of health management systems, and in particular to a comprehensive health management modeling system and method based on visual management. Background Art

[0002] With the integration of medical technology and information technology, health management has become the focus of attention. Traditional health management methods usually rely on the professional knowledge and experience of doctors, but with the growth of data volume and complexity, this method can no longer meet the needs of personalized health management. Existing health management solutions include regular physical examinations, health consultations, exercise and fitness plans, and diet management. However, existing solutions can only provide single-aspect health management. For example, physical examinations only focus on physiological indicators, and fitness plans only focus on exercise. There is a lack of comprehensive data integration, and most solutions are based on the experience and suggestions of the general population. There is a lack of in-depth analysis and personalized suggestions for individuals, which makes it difficult for users to track and manage their health data in a long-term and continuous manner, so the effect is limited.

[0003] For example, the patent with publication number CN110600093A discloses a health management method, system, terminal device and storage medium, including obtaining the health diagnosis results of the selected person; storing the health diagnosis results in the health file of the selected person; generating a health status analysis report based on the health record information within a preset time range in the health file, the health record information includes the health diagnosis results; generating health management suggestions based on the health status analysis report. Through the above process, the user's physical information can be controlled in real time, and the health management suggestions can be changed at any time according to the real-time physical information, effectively protecting the user's physical health.

[0004] For example, a patent with authorization announcement number CN114708941B discloses a management method and device based on health data, which includes: sending at least one original health data of a user obtained by an input party to a data processing system; the data processing system optimizes the original health data, obtains the processed health data corresponding to the user and sends it to a data management platform; the data management platform performs data modeling on the processed health data, obtains a star model of the user's logical data and performs data classification on the data in the star model to obtain health data of different levels corresponding to the user, configures different data application management and control strategies for health data of different levels, and outputs the corresponding health data of different levels to the data application side based on different data application management and control strategies, thereby improving the security and usability of health data as well as the convenience and efficiency of using health data.

[0005] The above existing technologies all have the following problems: 1) only provide single-aspect health management; 2) classify data and configure different data application management and control strategies according to different levels, resulting in abuse, misuse or omission of data. Summary of the invention

[0006] In view of the shortcomings of the existing technology, the present invention proposes a comprehensive health management modeling method based on visual management. The physiological, activity and dietary data of users are collected through automated means. After pre-processing, the survival analysis strategy is used to identify health risks. A risk assessment model is constructed by combining machine learning and statistical methods. The abnormal data is judged in real time. For abnormal data, natural language processing technology is used to provide feedback to the user. For health data, causal reasoning is used to predict future health status. Then, personalized health management plans are formulated based on user data, genetic information and environmental factors. The effectiveness of the plan is evaluated through non-invasive monitoring, and dynamic adjustments are made according to the evaluation results to ensure the real-time and effectiveness of health management.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A comprehensive health management modeling method based on visual management, including:

[0009] Step S1: Obtain the user's physiological data, activity data, and dietary data through an automated collection method, perform preprocessing, and identify health risks and abnormalities based on a survival analysis strategy;

[0010] Step S2: Based on the survival analysis results, a risk assessment model is constructed by combining machine learning algorithms and statistical methods to evaluate the user's health status and risk level in real time;

[0011] Step S3: Based on the results of the risk assessment model, determine whether the input data is abnormal data. If it is abnormal data, introduce natural language processing technology according to the type and severity of the abnormal data, and present the abnormal detection results to the user in the form of natural language. If it is health data, use causal reasoning methods to identify the causes of health results, and predict future health status through predictive analysis;

[0012] Step S4: Set the initial weights of physiological data, activity data, and dietary data, introduce an adaptive learning strategy, automatically formulate a behavioral intervention strategy based on the user's health status, risk prediction, and feedback needs, dynamically adjust the weight relationship between physiological data, activity data, and dietary data, and formulate a personalized health management plan in combination with the user's genetic information and environmental factors;

[0013] Step S5: Monitor the user's health status and the implementation of the health management plan through non-invasive strategies, evaluate the health management effect, and adjust the health management plan in a timely manner according to the evaluation results.

[0014] Specifically, the specific steps of the survival analysis strategy in step S1 include:

[0015] S1.1: Construct a survival analysis model and input the preprocessed data X = {a, b, c} into the survival analysis model for model fitting, where a = {a1, ..., a k} represents physiological data, a k represents the kth physiological data, k represents the number of physiological data, b={b1,…,b m} represents activity data, b m represents the mth activity data, m represents the number of activity data, c={c1,…,c n} represents dietary data, c n represents the nth dietary data, where n represents the number of dietary data;

[0016] S1.2: Calculate the health risk incidence and survival function estimate at each time point, draw the survival curve, and verify the survival analysis model to check whether the basic assumptions of survival analysis are met. The formula is:

[0017]

[0018] Among them, S t represents the cumulative survival rate, t represents the time distribution from healthy state to disease state, and t i represents the time when the i-th health risk occurs, x1 represents the number of health risks occurring at time t1, y1 represents the size of the risk set at time t1, x2 represents the number of health risks occurring at time t2, y2 represents the size of the risk set at time t2, x i Indicates that at time t i Number of health risks that occurred, y i Indicates that at time t i The size of the risk set at the time

[0019] S1.3: Set the significance level threshold δ = 0.05, based on the survival curve results, and obtain the significance level p of each covariate according to the log-rank test statistic. If p < δ, the covariate is significantly correlated with the survival time or the incidence of health risks. Traverse to obtain all covariates X1 = {a′, b′, c′} that affect the survival time or the incidence of health risks, where a′ represents the physiological data covariate, b′ represents the activity data covariate, and c′ represents the dietary data covariate;

[0020] S1.4: Explain and visualize identified health risks and anomalies.

[0021] Specifically, the specific steps of calculating the health risk incidence rate at each time point in S1.2 include:

[0022] S1.21: According to the output results of the survival analysis model, the survival time and survival status of each input data at that time are obtained, and the data are arranged in ascending order according to the survival time;

[0023] S1.22: Set the health risk incidence rate P at each time point f Initialize to 0, traverse the data in ascending order, and for each time point, calculate the proportion of health risks among all individuals before that time point. The formula is:

[0024]

[0025] Among them, z t Indicates the cumulative number of data on health risks that have occurred, y t It represents the total number of data that are still in the risk set before time point t;

[0026] S1.23: Draw a survival curve, where the horizontal axis represents time and the vertical axis represents cumulative survival rate, and add confidence intervals and risk tables to the survival curve.

[0027] Specifically, the specific steps of step S2 include:

[0028] S2.1: Use the preprocessed data X = {a, b, c} to train the selected machine learning model and evaluate the model performance through cross-validation;

[0029] S2.2: Input the results of survival analysis as features into the machine learning model and perform ensemble learning;

[0030] S2.3: Based on the results of the machine learning model, the user's health data is collected in real time, and the updated data X2 = {a″, b″, c″} is input into the trained risk assessment model to obtain real-time risk assessment results, where a″ represents the updated physiological data, b″ represents the updated activity data, and c″ represents the updated dietary data.

[0031] Specifically, the specific steps of step S3 include:

[0032] S3.1: Set the risk score threshold T, and determine whether the input data X2 = {a″, b″, c″} belongs to the abnormal data range according to the results of the risk assessment model;

[0033] S3.2: If X2(:)>T, the data is considered abnormal data, X2(:) represents any value in the data X2={a″,b″,c″}, and the abnormal data is classified according to abnormal physiological indicators, abnormal activities, abnormal diet and the P corresponding to each data. f The values ​​are classified and the anomaly detection results are visualized using natural language processing technology.

[0034] S3.3: If X2(:)≤T, the data is considered as health data, and the causal reasoning method is used to obtain health results.

[0035] Specifically, the behavioral intervention strategies in step S4 include: dietary adjustment suggestions, exercise plans, lifestyle improvement suggestions, health monitoring and tracking, education and training.

[0036] Specifically, the non-invasive strategy in step S5 includes: physiological parameter monitoring, activity level monitoring, sleep monitoring, diet record and analysis, questionnaire survey and self-assessment.

[0037] A comprehensive health management modeling system based on visual management, comprising:

[0038] Data collection module, health analysis module, health management module, visualization module, multimodal interaction module;

[0039] The data collection module is used to collect physiological data, activity data and dietary data;

[0040] The health analysis module is used to analyze and evaluate the collected health data based on survival analysis strategies, machine learning algorithms and statistical methods, and generate information about health conditions and risks;

[0041] The health management module is used to provide dietary advice, exercise plans, and behavioral intervention strategies based on the results of the health analysis module, and to make dynamic adjustments;

[0042] The visualization display module is used to visualize the health analysis and management results in an intuitive and dynamic manner;

[0043] The multimodal interaction module is used to provide voice interaction, gesture recognition, and touch screen interaction methods, so that users can interact with the system.

[0044] Specifically, the health analysis module includes: a survival analysis unit and a risk assessment unit;

[0045] The survival analysis unit is used to estimate the probability of the user remaining healthy or experiencing a specific health event within a given time period based on the user's health data;

[0046] The risk assessment unit is used to analyze the user's health status, identify potential health problems, and predict health risks that may occur in the future.

[0047] Specifically, the health management module includes: a goal setting unit, a prediction unit, a plan making unit, and a progress tracking unit;

[0048] The goal setting unit is used to help users clarify and set health management goals;

[0049] The plan making unit predicts the future health status based on the user's health data, historical performance and health management goals;

[0050] The prediction unit is used to formulate a health management plan for the user according to the user's health management goals and prediction results;

[0051] The progress tracking unit is used to monitor the implementation of the user's health management plan and provide timely feedback to the user and the health manager.

[0052] Compared with the prior art, the present invention has the following beneficial effects:

[0053] 1. The present invention proposes a comprehensive health management modeling system based on visual management, and optimizes and improves the architecture, operation steps and processes. The system has the advantages of simple process, low investment and operation costs, and low production work costs. At the same time, it combines the user's genetic information, environmental factors, and real-time collected physiological, activity and dietary data to provide each user with a personalized health management plan, which can better meet the actual needs of users than traditional "one-size-fits-all" health advice and improve the effect of health management.

[0054] 2. The present invention proposes a comprehensive health management modeling method based on visual management. By introducing machine learning algorithms and statistical methods, the user's health status and risk level are evaluated in real time, and the future health status is predicted. The results are presented to the user through natural language processing technology, which helps the user to promptly discover and deal with health problems and prevent the problems from worsening. The non-invasive strategy is used to monitor the user's health status and the implementation of the health management plan, which can comprehensively evaluate the health management effect, discover problems in time and make adjustments. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 This is a flow chart of a comprehensive health management modeling method based on visual management of the present invention;

[0056] Figure 2 This is a schematic diagram of the overall process of a comprehensive health management modeling method based on visual management of the present invention;

[0057] Figure 3This is an architecture diagram of a comprehensive health management modeling system based on visual management in the present invention. DETAILED DESCRIPTION

[0058] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, in the description of the present invention, it should be noted that the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside" and the like indicate the orientation or position relationship based on the orientation or position relationship shown in the accompanying drawings, which is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. In addition, the terms "No. 1", "No. 2" and "No. 3" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. The present invention is further explained below in conjunction with specific embodiments.

[0059] Example 1

[0060] See also Figure 1-Figure 2 , an embodiment provided by the present invention: a comprehensive health management modeling method based on visual management, comprising the following steps:

[0061] Step S1: Obtain the user's physiological data, activity data, and dietary data through an automated collection method, perform preprocessing, and identify health risks and abnormalities based on a survival analysis strategy;

[0062] Preprocessing includes: data cleaning, verification and data fusion.

[0063] The present invention uses wearable devices and smart home devices to realize the automatic collection of physiological data, activity data and dietary data, reducing user participation and errors.

[0064] The present invention mainly includes: identifying health risks and abnormal conditions according to the survival analysis strategy:

[0065] (1) Guiding data collection and processing: In step S1, the survival analysis strategy is used to identify health risks and abnormal conditions, which can determine which physiological, activity, and dietary data are critical, thereby guiding subsequent data collection and processing to ensure that the collected data is valuable for risk identification and analysis;

[0066] (2) Constructing a risk assessment model: In step S2, based on the results of survival analysis, a more accurate risk assessment model can be constructed by combining machine learning algorithms and statistical methods;

[0067] (3) Determine data anomalies and formulate intervention strategies: In step S3, based on the results of the risk assessment model, it can be determined whether the input data is abnormal data. In this process, the survival analysis strategy can help identify which abnormal data are related to specific health risks or abnormal conditions, thereby formulating more accurate behavioral intervention strategies;

[0068] (4) Formulation of personalized health management plan: In step S4, when setting the initial weights of physiological, activity and dietary data, the results of survival analysis can be used as an important basis for weight setting. When introducing the adaptive learning strategy, the results of survival analysis can provide a basis for dynamically adjusting the data weights to ensure that the personalized health management plan is more in line with the user's actual health status and risk prediction. At the same time, combined with the user's genetic information and environmental factors, a more accurate and personalized health management plan can be formulated;

[0069] (5) Evaluation and adjustment of health management effectiveness: Through the survival analysis strategy, the effectiveness of the health management plan can be regularly evaluated, and the plan can be adjusted in a timely manner based on the evaluation results, which helps to ensure that the health management plan always keeps pace with the user's health status and risk prediction, thereby improving the effectiveness of health management.

[0070] In summary, the step of “identifying health risks and abnormal conditions based on survival analysis strategies” plays a key guiding role in the entire process of the comprehensive health management approach, and provides an important basis for subsequent data collection, processing, risk assessment, intervention strategy formulation, and the formulation and adjustment of personalized health management plans.

[0071] Step S2: Based on the survival analysis results, a risk assessment model is constructed by combining machine learning algorithms and statistical methods to evaluate the user's health status and risk level in real time;

[0072] The output results of the survival analysis model include the coefficient of each covariate, the standard error of these coefficients, the significance level, and the confidence interval statistics. These statistics reflect the relationship between the covariate and survival time or the incidence of health risks.

[0073] Step S3: Based on the results of the risk assessment model, determine whether the input data is abnormal data. If it is abnormal data, introduce natural language processing technology according to the type and severity of the abnormal data, and present the abnormal detection results to the user in the form of natural language. If it is health data, use causal reasoning methods to identify the causes of health results, and predict future health status through predictive analysis;

[0074] The introduction of natural language processing technology is to enable users to interact with the visual interface through voice commands and improve the user experience.

[0075] Step S4: Set the initial weights of physiological data, activity data, and dietary data, introduce an adaptive learning strategy, automatically formulate a behavioral intervention strategy based on the user's health status, risk prediction, and feedback needs, dynamically adjust the weight relationship between physiological data, activity data, and dietary data, and formulate a personalized health management plan in combination with the user's genetic information and environmental factors;

[0076] Physiological data, such as heart rate, blood pressure, blood sugar, and sleep quality; activity data, such as exercise volume, activity intensity, and daily activity volume; dietary data, such as food types, intake, and nutritional content.

[0077] Step S5: Monitor the user's health status and the implementation of the health management plan through non-invasive strategies, comprehensively evaluate the health management effect, and adjust the health management plan in a timely manner based on the evaluation results.

[0078] The comprehensive health management method used in the present invention mainly manages the overall health status of an individual, including physical, psychological, social adaptation and other aspects, aiming to help individuals prevent diseases, promote health and improve the quality of life through a series of comprehensive intervention measures. In the comprehensive health management method, health management refers to a comprehensive, personalized, preventive health service, which provides targeted health guidance and intervention measures by conducting a comprehensive assessment of the health status of an individual, helping the individual to establish a healthy lifestyle, reduce health risks and improve health levels. Health management services include health consultation, health risk assessment, health education, health intervention, and health effect evaluation, aiming to help individuals achieve comprehensive management of their physical and mental health.

[0079] Data fusion uses multimodal data fusion to integrate and analyze data from different sources, such as physiological data, activity data, and dietary data, to obtain a more comprehensive health status assessment.

[0080] The specific steps of the survival analysis strategy in step S1 include:

[0081] S1.1: Construct a survival analysis model and input the preprocessed data X = {a, b, c} into the survival analysis model for model fitting, where a = {a1,…, a k} represents physiological data, a k represents the kth physiological data, k represents the number of physiological data, b={b1,…,b m} represents activity data, b m represents the mth activity data, m represents the number of activity data, c={c1,…,c n} represents dietary data, c n represents the nth dietary data, where n represents the number of dietary data;

[0082] S1.2: Calculate the health risk incidence and survival function estimate at each time point, draw the survival curve, and verify the survival analysis model to check whether the basic assumptions of survival analysis are met. The formula is:

[0083]

[0084] Among them, S t represents the cumulative survival rate, which describes the probability that an individual is still alive at a given time point, t represents the time distribution of transition from a healthy state to a disease state, and t i represents the time when the i-th health risk occurs, x1 represents the number of health risks occurring at time t1, y1 represents the size of the risk set at time t1, x2 represents the number of health risks occurring at time t2, y2 represents the size of the risk set at time t2, x i Indicates that at time t i Number of health risks that occurred, y i Indicates that at time t i The size of the risk set at the time

[0085] The size of the risk set refers to the number of individuals for whom an event has not yet occurred. The incidence of health risks at each time point is also called the event incidence rate, which refers to the probability of an event occurring in the future instant for individuals who have survived to that time point at a given time point. If the incidence of health risks increases over time, it means that the health risk of individuals is increasing over time. On the contrary, if the incidence of health risks decreases over time, it may mean that the health status of individuals is improving or that certain intervention measures have taken effect. Among them, health risk refers to a risk that acts on the human body and affects human health among the existing risks. Specifically, health risk refers to the possibility of illness, disability and health loss in people due to many factors in nature, society and human development during the life process of people.

[0086] S1.3: Set the significance level threshold δ = 0.05, based on the survival curve results, and obtain the significance level p of each covariate according to the log-rank test statistic. If p < δ, the covariate is significantly correlated with the survival time or the incidence of health risks. Traverse to obtain all covariates X1 = {a′, b′, c′} that affect the survival time or the incidence of health risks, where a′ represents the physiological data covariate, b′ represents the activity data covariate, and c′ represents the dietary data covariate;

[0087] S1.4: Explain and visualize identified health risks and anomalies.

[0088] The specific steps for calculating the health risk incidence rate at each time point in S1.2 include:

[0089] S1.21: According to the output results of the survival analysis model, the survival time and survival status of each input data at that time are obtained, and the data are arranged in ascending order according to the survival time;

[0090] S1.22: Set the health risk incidence rate P at each time point f Initialize to 0, traverse the data in ascending order, and for each time point, calculate the proportion of health risks among all individuals before that time point. The formula is:

[0091]

[0092] Among them, z t Indicates the cumulative number of data on health risks that have occurred, y t It represents the total number of data that are still in the risk set before time point t;

[0093] S1.23: Draw a survival curve, where the horizontal axis represents time and the vertical axis represents cumulative survival rate, and add confidence intervals and risk tables to the survival curve.

[0094] The specific steps of step S2 include:

[0095] S2.1: Use the preprocessed data X = {a, b, c} to train the selected machine learning model and evaluate the model performance through cross-validation;

[0096] S2.2: Input the results of survival analysis as features into the machine learning model and perform ensemble learning;

[0097] S2.3: Based on the results of the machine learning model, the user's health data is collected in real time, and the updated data X2 = {a″, b″, c″} is input into the trained risk assessment model to obtain real-time risk assessment results, where a″ represents the updated physiological data, b″ represents the updated activity data, and c″ represents the updated dietary data.

[0098] The specific steps of step S3 include:

[0099] S3.1: Set the risk score threshold T, and determine whether the input data X2 = {a″, b″, c″} belongs to the abnormal data range according to the results of the risk assessment model;

[0100] S3.2: If X2(:)>T, the data is considered abnormal data, X2(:) represents any value in the data X2={a″,b″,c″}, and the abnormal data is classified according to abnormal physiological indicators, abnormal activities, abnormal diet and the P corresponding to each data. f The values ​​are classified and the anomaly detection results are visualized using natural language processing technology.

[0101] S3.3: If X2(:)≤T, the data is considered as health data, and the causal reasoning method is used to obtain health results.

[0102] The behavioral intervention strategies in step S4 include: dietary adjustment suggestions, exercise plans, lifestyle improvement suggestions, health monitoring and tracking, education and training.

[0103] The non-invasive strategies in step S5 include: physiological parameter monitoring, activity level monitoring, sleep monitoring, diet record and analysis, questionnaire survey and self-assessment.

[0104] The non-invasive strategy mainly uses spectral analysis and acoustic wave technology to evaluate the health status of organisms and achieve continuous monitoring of the user's health status.

[0105] Example 2

[0106] See also Figure 3 Another embodiment provided by the present invention is a comprehensive health management modeling system based on visual management, comprising:

[0107] Data collection module, health analysis module, health management module, visualization module, multimodal interaction module;

[0108] The data collection module is used to collect physiological data, activity data and dietary data;

[0109] The health analysis module is used to analyze and evaluate the collected health data based on survival analysis strategies, machine learning algorithms and statistical methods, and generate information about health conditions and risks;

[0110] The health management module is used to provide dietary advice, exercise plans, and behavioral intervention strategies based on the results of the health analysis module, and to make dynamic adjustments;

[0111] The visualization display module is used to visualize the health analysis and management results in an intuitive and dynamic manner;

[0112] The multimodal interaction module is used to provide voice interaction, gesture recognition, and touch screen interaction methods, so that users can interact with the system.

[0113] In addition to the data collection module, health analysis module, health management module, visualization display module, and multimodal interaction module, a comprehensive health management modeling system based on visual management can also introduce predictive health modules, emotional intelligence modules, intelligent exercise guidance modules, nutrition recommendation and diet management modules, health ecosystem integration modules, health education and knowledge popularization modules, and intelligent drug management modules to provide users with more comprehensive, personalized, and intelligent health management services.

[0114] The health analysis module includes: a survival analysis unit and a risk assessment unit;

[0115] The survival analysis unit is used to estimate the probability of the user remaining healthy or experiencing a specific health event within a given time period based on the user's health data;

[0116] The risk assessment unit is used to analyze the user's health status, identify potential health problems, and predict health risks that may occur in the future.

[0117] The health management module includes: a goal setting unit, a prediction unit, a plan making unit, and a progress tracking unit;

[0118] The goal setting unit is used to help users clarify and set health management goals;

[0119] The plan making unit predicts the future health status based on the user's health data, historical performance and health management goals;

[0120] The prediction unit is used to formulate a specific health management plan for the user according to the user's health management goals and prediction results;

[0121] The progress tracking unit is used to monitor the implementation of the user's health management plan and provide timely feedback to the user and the health manager.

[0122] The visualization display module includes: a chart display unit, which displays health data in the form of charts to facilitate user understanding; a data comparison unit, which compares the user's health data in different time periods and displays the changing trend; a health report unit, which generates the user's health report for the user to view at any time.

[0123] The multimodal interaction module includes: an interface design unit, which designs a friendly and easy-to-use user interface; a feedback collection unit, which collects user feedback on the system and continuously optimizes system performance; and an education support unit, which provides health education materials to help users enhance their health awareness.

[0124] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in the field may also change, modify, replace and modify the above-mentioned embodiments without departing from the scope of protection of the purpose of the present invention and the claims, and all of these are within the protection of the present invention.

Claims

1. A comprehensive health management modeling method based on visual management, characterized in that: include: Step S1: Obtain the user's physiological data, activity data, and dietary data through an automated collection method, perform preprocessing, and identify health risks and abnormalities based on a survival analysis strategy; Step S2: Based on the survival analysis results, a risk assessment model is constructed by combining machine learning algorithms and statistical methods to evaluate the user's health status and risk level in real time; Step S3: Based on the results of the risk assessment model, determine whether the input data is abnormal data. If it is abnormal data, introduce natural language processing technology according to the type and severity of the abnormal data, and present the abnormal detection results to the user in the form of natural language. If it is health data, use causal reasoning methods to identify the causes of health results, and predict future health status through predictive analysis; Step S4: Set the initial weights of physiological data, activity data, and dietary data, introduce an adaptive learning strategy, automatically formulate a behavioral intervention strategy based on the user's health status, risk prediction, and feedback needs, dynamically adjust the weight relationship between physiological data, activity data, and dietary data, and formulate a personalized health management plan in combination with the user's genetic information and environmental factors; Step S5: Monitor the user's health status and the implementation of the health management plan through non-invasive strategies, evaluate the health management effect, and adjust the health management plan in a timely manner according to the evaluation results.

2. A comprehensive health management modeling method based on visual management as claimed in claim 1, characterized in that: The specific steps of the survival analysis strategy in step S1 include: S1.1: Construct a survival analysis model and input the preprocessed data X = {a, b, c} into the survival analysis model for model fitting, where a = {a1, ..., a k } represents physiological data, a k represents the kth physiological data, k represents the number of physiological data, b={b1,...,b m } represents activity data, b m represents the mth activity data, m represents the number of activity data, c={c1,...,c n } represents dietary data, c n represents the nth dietary data, where n represents the number of dietary data; S1.2: Calculate the health risk incidence and survival function estimate at each time point, draw the survival curve, and verify the survival analysis model to check whether the basic assumptions of survival analysis are met. The formula is: Among them, S t represents the cumulative survival rate, t represents the time distribution from healthy state to disease state, and t i represents the time when the i-th health risk occurs, x1 represents the number of health risks occurring at time t1, y1 represents the size of the risk set at time t1, x2 represents the number of health risks occurring at time t2, y2 represents the size of the risk set at time t2, x i Indicates that at time t i Number of health risks that occurred, y i Indicates that at time t i The size of the risk set at the time S1.3: Set the significance level threshold δ = 0.05, based on the survival curve results, and obtain the significance level p of each covariate according to the log-rank test statistic. If p < δ, the covariate is significantly correlated with the survival time or the incidence of health risks. Traverse to obtain all covariates X1 = {a′, b′, c′} that affect the survival time or the incidence of health risks, where a′ represents the physiological data covariate, b′ represents the activity data covariate, and c′ represents the dietary data covariate; S1.4: Explain and visualize identified health risks and anomalies.

3. A comprehensive health management modeling method based on visual management as claimed in claim 2, characterized in that: The specific steps for calculating the health risk incidence rate at each time point in S1.2 include: S1.21: According to the output results of the survival analysis model, the survival time and survival status of each input data at that time are obtained, and the data are arranged in ascending order according to the survival time; S1.22: Set the health risk incidence rate P at each time point f Initialize to 0, traverse the data in ascending order, and for each time point, calculate the proportion of health risks among all individuals before that time point. The formula is: Among them, z t Indicates the cumulative number of data on health risks that have occurred, y t It represents the total number of data that are still in the risk set before time point t; S1.23: Draw a survival curve, where the horizontal axis represents time and the vertical axis represents cumulative survival rate, and add confidence intervals and risk tables to the survival curve.

4. A comprehensive health management modeling method based on visual management as claimed in claim 3, characterized in that: The specific steps of step S2 include: S2.1: Use the preprocessed data X = {a, b, c} to train the selected machine learning model and evaluate the model performance through cross-validation; S2.2: Input the results of survival analysis as features into the machine learning model and perform ensemble learning; S2.3: Based on the results of the machine learning model, the user's health data is collected in real time, and the updated data X2 = {a″, b″, c″} is input into the trained risk assessment model to obtain real-time risk assessment results, where a″ represents the updated physiological data, b″ represents the updated activity data, and c″ represents the updated dietary data.

5. A comprehensive health management modeling method based on visual management as claimed in claim 4, characterized in that: The specific steps of step S3 include: S3.1: Set the risk score threshold T, and determine whether the input data X2 = {a″, b″, c″} belongs to the abnormal data range according to the results of the risk assessment model; S3.2: If X2(:)>T, the data is considered abnormal data, X2(:) represents any value in the data X2={a″,b″,c″}, and the abnormal data is classified according to abnormal physiological indicators, abnormal activities, abnormal diet and the P corresponding to each data. f The values ​​are classified and the anomaly detection results are visualized using natural language processing technology. S3.3: If X2(:)≤T, the data is considered as health data, and the causal reasoning method is used to obtain the health results.

6. A comprehensive health management modeling method based on visual management as claimed in claim 5, characterized in that: The behavioral intervention strategies in step S4 include: dietary adjustment suggestions, exercise plans, lifestyle improvement suggestions, health monitoring and tracking, education and training.

7. A comprehensive health management modeling method based on visual management as claimed in claim 6, characterized in that: The non-invasive strategies in step S5 include: physiological parameter monitoring, activity level monitoring, sleep monitoring, diet record and analysis, questionnaire survey and self-assessment.

8. A comprehensive health management modeling system based on visual management, the system is used to implement a comprehensive health management modeling method based on visual management according to any one of claims 1 to 7, characterized in that: include: Data collection module, health analysis module, health management module, visualization module, multimodal interaction module; The data collection module is used to collect physiological data, activity data and dietary data; The health analysis module is used to analyze and evaluate the collected health data based on survival analysis strategies, machine learning algorithms and statistical methods, and generate information about health conditions and risks; The health management module is used to provide dietary advice, exercise plans, and behavioral intervention strategies based on the results of the health analysis module, and to make dynamic adjustments; The visualization display module is used to visualize the health analysis and management results in an intuitive and dynamic manner; The multimodal interaction module is used to provide voice interaction, gesture recognition, and touch screen interaction methods, so that users can interact with the system.

9. A comprehensive health management modeling system based on visual management as claimed in claim 8, characterized in that: The health analysis module includes: a survival analysis unit and a risk assessment unit; The survival analysis unit is used to estimate the probability of the user remaining healthy or experiencing a specific health event within a given time period based on the user's health data; The risk assessment unit is used to analyze the user's health status, identify potential health problems, and predict health risks that may occur in the future.

10. A comprehensive health management modeling system based on visual management as claimed in claim 9, characterized in that: The health management module includes: a goal setting unit, a prediction unit, a plan making unit, and a progress tracking unit; The goal setting unit is used to help users clarify and set health management goals; The plan making unit predicts the future health status based on the user's health data, historical performance and health management goals; The prediction unit is used to formulate a health management plan for the user according to the user's health management goals and prediction results; The progress tracking unit is used to monitor the implementation of the user's health management plan and provide timely feedback to the user and the health manager.

Images