Insurance risk assessment and health management scheme generation system for sub-health population

Abstract

The invention provides an insurance risk assessment and health management scheme generation system for sub-health crowds, which is applied to the technical field of health management, an acquisition module is responsible for collecting actual data of a user in the process of executing a health management scheme, and a modeling module quantifies the physiological response degree of an individual to the scheme and outputs a physiological response value. And the evaluation module evaluates the overall effect of the health management scheme currently executed by the user by using the physiological response value output by the modeling module. And the adjustment module modifies the health management scheme of the user according to the quantitative evaluation result obtained by the evaluation module. Through a cycle process of acquisition, modeling, evaluation and adjustment, the system can continuously monitor individual response and dynamically optimize the scheme, so that the actual improvement effect of the scheme on individual health is improved, and a data basis is provided for more accurate insurance risk evaluation.

Description

Technical Field

[0001] The present application relates to the field of health management technology, and in particular to a system for generating insurance risk assessment and health management plans for sub-healthy people. Background Art

[0002] Sub-health is a dynamic process between health and disease, characterized by various physical symptoms that do not meet the diagnostic criteria for a clinical disease. Effective health management for sub-healthy individuals not only helps improve individual health and prevent disease, but also enables the insurance industry to more accurately assess and manage individual health risks, thereby optimizing insurance product design, underwriting processes, and pricing strategies.

[0003] Existing health management systems for people with sub-health conditions typically focus on conducting individual insurance risk assessments and generating health management plans. Once the risk assessment is complete, the system generates a personalized health management plan based on the assessment results, the user's current health status, and the preferences expressed when filling out the information. This plan may include dietary recommendations, exercise plans, routine adjustment suggestions, stress management techniques, and other aspects. The plan generation process attempts to consider the user's actual situation and preferences to enhance feasibility.

[0004] However, in actual applications, it is found that the effects of such health management programs on specific individuals vary significantly. For example, the system may recommend a certain type of exercise based on the user's physical indicators, but for users with specific physiological conditions (such as joint problems, metabolic characteristics) or in specific environmental factors (such as high work pressure, lack of sleep), the actual health improvement effect of the exercise program may be far lower than expected. This means that the system fails to fully understand the key factors that affect whether specific interventions can produce the expected effects on specific individuals, resulting in the generated health management program may not be the most suitable and effective path for the user.

[0005] Therefore, in a system that serves sub-healthy people and aims to conduct insurance risk assessment and health management, facing the problem that existing health management programs are difficult to capture the differences in individual physiological responses to specific interventions, resulting in uncertain program effects and risk assessment bias, how to continuously monitor the dynamic data of individual physiological indicators of users during the implementation of health management programs, build a mechanism that can quantify the correlation between changes in individual physiological indicators and specific health interventions, and dynamically evaluate the effectiveness of the program and adjust the program based on this quantitative result, so as to enhance the actual improvement effect of the program on individual health and achieve accurate management of risk assessment. Summary of the Invention

[0006] In view of the above-mentioned shortcomings of the existing technology, the present application provides an insurance risk assessment and health management plan generation system for sub-healthy people, which is applied in the field of health management technology and has the advantages of quantifying individual physiological responses, evaluating the effectiveness of plans, adjusting health management plans, improving the effectiveness of plans, and realizing accurate risk assessment management.

[0007] In a first aspect, a system for insurance risk assessment and health management plan generation for sub-healthy people is provided, the system comprising:

[0008] Collection module: collects individual physiological indicator data and plan execution information of users during the implementation of health management plans;

[0009] Modeling module: receiving the individual physiological indicator data and the program execution information, analyzing the dynamic change trend of the individual physiological indicator data after the specific program is executed in the program execution information, and establishing a correlation model between the individual physiological indicator changes and the specific program to quantify the individual's physiological response to the specific program and obtain a physiological response value;

[0010] Evaluation module: evaluates the overall effect of the current health management program on the user based on the physiological response value and obtains a quantitative evaluation result;

[0011] Adjustment module: adjust the user's health management plan according to the quantitative evaluation results.

[0012] This application proposes a system for generating insurance risk assessment and health management plans for sub-healthy people. Through a cyclical process of collection, modeling, evaluation, and adjustment, the system can continuously monitor individual responses and dynamically optimize plans, thereby enhancing the actual improvement effect of the plans on individual health and providing a data basis for more accurate insurance risk assessment.

[0013] Furthermore, the modeling module includes:

[0014] Receiving module: receiving the individual physiological indicator data and the program execution information, and preprocessing the individual physiological indicator data to eliminate abnormal values ​​and missing values ​​to obtain valid physiological indicator data;

[0015] Extraction module: using a time series analysis method to analyze the dynamic change trend of the effective physiological indicator data before and after the execution of the specific plan, and extracting the change amplitude, change rate and duration as physiological response characteristics;

[0016] Modeling submodule: taking the information of the execution of the specific plan as input and the physiological response characteristics as output, establishing a correlation model between the changes in individual physiological indicators and the specific plan, and obtaining the physiological response coefficient corresponding to the specific plan;

[0017] Calculation module: Calculates the individual's physiological response degree to the specific program based on the physiological response coefficient to obtain a quantified physiological response value.

[0018] This application proposes a system for generating insurance risk assessment and health management plans for sub-healthy people. Through the collaborative work of the above modules, this technical solution provides a specific and operational modeling process, from data cleaning and feature extraction to model establishment and result calculation, which improves the accuracy and reliability of quantifying individual physiological responses and lays the foundation for subsequent plan evaluation and adjustment.

[0019] Furthermore, the information of specific program execution includes exercise type, exercise intensity, exercise duration, food intake, and food type; the modeling submodule includes:

[0020] The first modeling module: construct a linear regression model including L1 regularization term as the association model between changes in individual physiological indicators and specific plans;

[0021] The second modeling module: based on the historical health management data, optimizes the L1 regularization coefficient of the linear regression model to obtain an optimized linear regression model;

[0022] The third modeling module: One-hot encoding is performed on the exercise type and food type, and the encoded exercise type, food type, exercise intensity, exercise duration, and food intake are input into the optimized linear regression model to calculate the physiological response coefficient corresponding to the specific plan.

[0023] This application proposes a system for generating insurance risk assessment and health management plans for sub-healthy people. By adopting and optimizing the L1 regularized linear regression model and combining it with encoding processing of classification input, this technical solution can more effectively learn individual physiological response patterns from multi-dimensional plan information, identify key influencing factors, and thus establish a more accurate association model and calculate physiological response coefficients that can better reflect individual differences, solving the problem that existing methods are difficult to accurately quantify individual physiological responses.

[0024] Furthermore, the third modeling module includes:

[0025] Encoding module: performing one-hot encoding on the exercise type and the food type to obtain corresponding vector codes;

[0026] Integration module: splicing and integrating the vector code with the values ​​of exercise intensity, exercise duration, and food intake to obtain a feature vector of a specific plan;

[0027] The first calculation module is to input the characteristic vector into the optimized linear regression model to calculate the physiological response coefficient corresponding to the specific solution.

[0028] This application proposes a system for generating insurance risk assessment and health management plans for sub-healthy people. It converts non-numerical categories into numerical forms that can be processed by the model by introducing an encoding module, integrates different types of data into a unified input vector through an integration module, and obtains calculation results through a first calculation module.

[0029] Furthermore, the first calculation module includes:

[0030] A second calculation module: inputs the feature vector into the optimized linear regression model and outputs a physiological response prediction value;

[0031] The third calculation module: uses the formula: physiological impact coefficient = (physiological impact prediction value * standard deviation of physiological indicator data) + mean value of physiological indicator data to calculate the physiological impact coefficient corresponding to the specific plan.

[0032] Furthermore, the assessment modules include:

[0033] Acquisition module: Acquisition of a preset health management program evaluation index system, wherein the evaluation index system includes a plurality of physiological index threshold ranges related to sub-health and corresponding weight coefficients;

[0034] A fourth calculation module: comparing the physiological response value with the physiological indicator threshold range in the evaluation indicator system, determining the compliance status of each physiological indicator, and calculating the comprehensive compliance rate according to the corresponding weight coefficient;

[0035] The fifth calculation module: based on the comprehensive compliance rate, program implementation compliance and user feedback score, calculate the overall effect score of the current health management program on the user to obtain a quantitative evaluation result.

[0036] Furthermore, the fifth calculation module includes:

[0037] The sixth calculation module: obtains the user's plan execution records during the health management plan execution cycle, and calculates the plan execution compliance score based on the degree of match between the plan items actually executed and the recommended plan items in the records, where the plan execution compliance score = (the number of plan items actually executed / the total number of recommended plan items) * 100%;

[0038] The seventh calculation module collects feedback questionnaires filled out by users after the health management plan execution cycle. The questionnaires cover multiple dimensions such as satisfaction with the plan, ease of use, and perceived effectiveness. The questionnaire results are normalized to obtain user feedback scores.

[0039] The eighth calculation module: Using the weighted average method, the comprehensive compliance rate, program implementation compliance score and user feedback score are weighted and calculated to obtain the overall effect score of the current health management program on the user.

[0040] Furthermore, the adjustment module includes:

[0041] Prediction module: If the quantitative assessment result is lower than a preset effect threshold, the prediction module predicts the expected change in the user's physiological indicator after adjusting at least one of the exercise type, exercise intensity, exercise duration, food intake, and food type based on the correlation model between the change in individual physiological indicators and the specific plan;

[0042] Adjustment submodule: Select an adjustment plan that makes the expected change value of at least one key physiological indicator reach the preset target, and combine it with the preset plan adjustment rule library to make constrained adjustments to the user's health management plan, and generate the adjusted health management plan and push it to the user.

[0043] Furthermore, in the adjustment submodule, the constraint adjustment includes: ensuring that the adjusted exercise intensity does not exceed 80% of the user's maximum heart rate, and ensuring that the adjusted food intake does not exceed 120% of the user's total daily energy requirement.

[0044] Furthermore, the acquisition module includes:

[0045] Monitoring module: real-time monitoring and recording of the user's individual physiological indicators, including heart rate, blood pressure, blood sugar, weight, sleep duration, and activity level;

[0046] Encryption module: uploads individual physiological indicator data and program execution information to the cloud server through a data encryption channel. The program execution information includes exercise type, exercise intensity, exercise duration, food intake, and food type;

[0047] Storage module: receives the individual physiological indicator data and the program execution information, and writes the data into the database in the order of timestamp to ensure the time sequence and consistency of the data.

[0048] Beneficial effects: This application proposes a system for generating insurance risk assessment and health management plans for sub-healthy people. Through a cyclical process of collection, modeling, evaluation, and adjustment, the system can continuously monitor individual responses and dynamically optimize plans, thereby enhancing the actual improvement effect of the plans on individual health and providing a data basis for more accurate insurance risk assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 This is a structural diagram of the insurance risk assessment and health management plan generation system for sub-healthy people proposed in this application.

[0050] Figure 2 This is a structural diagram of the acquisition module proposed in this application.

[0051] Figure 3 Structural diagram of the evaluation module proposed for this application.

[0052] Figure 4 This is a structural diagram of the adjustment module proposed in this application. DETAILED DESCRIPTION

[0053] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application generally described and marked in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work fall within the scope of protection of the present application.

[0054] It should be noted that similar reference numerals and letters represent similar items in the following figures. Therefore, once an item is defined in one figure, it does not need to be further defined or explained in subsequent figures. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and should not be understood as indicating or implying relative importance.

[0055] Please refer to Figures 1 to 4 In the first aspect, a system for generating insurance risk assessment and health management plans for sub-healthy people is provided, the system comprising:

[0056] Collection module: collects individual physiological indicator data and plan execution information of users during the implementation of health management plans;

[0057] Modeling module: Receives individual physiological indicator data and program execution information, analyzes the dynamic change trend of individual physiological indicator data after the specific program is executed in the program execution information, and establishes a correlation model between the changes in individual physiological indicators and the specific program to quantify the individual's physiological response to the specific program and obtain the physiological response value;

[0058] Evaluation module: Evaluate the overall effect of the current health management program on the user based on physiological response values ​​and obtain quantitative evaluation results;

[0059] Adjustment module: Adjust the user's health management plan based on the quantitative evaluation results.

[0060] The acquisition module can be configured with sensor devices or data interfaces to automatically or manually acquire the user's heart rate, blood pressure, blood sugar, weight, sleep duration, activity level, and other physiological indicators, and record the user's actual exercise type, intensity, duration, food intake, food type, and other program execution information. This data is transmitted to the system for processing.

[0061] The modeling module receives the collected data and identifies the change patterns of physiological indicators before and after the implementation of the program through data processing and analysis algorithms, such as time series analysis.

[0062] Furthermore, the modeling module constructs a mathematical model, such as a regression model, taking the specific content of the program execution as input and the changing characteristics of the physiological indicators as output, and calculates the physiological response value reflecting the individual response degree.

[0063] The evaluation module receives the physiological response value and, in combination with the preset health assessment criteria or thresholds, calculates the quantitative score of the program's improvement in the individual's health status.

[0064] The adjustment module receives the evaluation results. If the evaluation score does not meet the target, it predicts the changes in physiological indicators that may be brought about by different adjustment directions of the plan based on the association model established by the modeling module, and generates a new health management plan based on the preset adjustment rules.

[0065] Specifically, the system aims to address the problem that existing health management programs fail to effectively capture individual differences in physiological responses to specific interventions, thereby improving program effectiveness and enhancing the accuracy of risk assessments. The system achieves this goal by establishing a closed-loop feedback mechanism.

[0066] First, the acquisition module continuously collects the user's physiological data and program execution details during the implementation of the health management program. These data are the basis for the system to conduct personalized analysis. Next, the modeling module receives this data and analyzes how the user's physiological indicators change over time after implementing a specific program (such as a certain exercise or diet adjustment). By constructing a correlation model between changes in physiological indicators and specific intervention measures, the system can quantify the individual's actual physiological response to the intervention and generate a physiological response value. The physiological response value directly reflects the actual effect of the intervention on the individual, overcoming the limitation of traditional programs that cannot quantify individual differences.

[0067] The evaluation module then uses this physiological response value, combined with other factors, to quantitatively assess the overall effectiveness of the health management program currently being implemented by the user. The evaluation results provide an objective indicator of whether the program is suitable for the individual.

[0068] Finally, the Adjustment Module dynamically modifies the user's health management plan based on the quantitative evaluation results from the Evaluation Module. If the evaluation shows that the current plan is not effective, the system can use the model established by the Modeling Module to predict the changes in physiological indicators that may be caused by the adjustment plan (such as changing exercise intensity or diet structure) and generate an optimized plan more suitable for the individual.

[0069] Through a continuous cycle of collection, modeling, evaluation, and adjustment, the system can continuously learn the individual's physiological response patterns and dynamically optimize health management plans, thereby enhancing the plan's actual improvement effect on individual health and providing data support for more accurate insurance risk assessment.

[0070] In some embodiments, the system is deployed on a cloud server, and users interact with the system through a mobile application or wearable device. The collection module receives physiological data streams from the user's device (e.g., hourly heart rate recording, daily weight recording) and user-entered program execution records (e.g., recording the type, duration, and intensity of each exercise session, or recording daily dietary intake) via an API interface.

[0071] After receiving this data, the modeling module cleans and preprocesses the physiological data, and then uses time series analysis methods to extract the changing characteristics of physiological indicators, such as calculating the changes in heart rate recovery rate after a certain exercise or the fluctuation range of blood sugar levels after diet adjustment.

[0072] Furthermore, the modeling module constructs a linear regression model, taking the program characteristics such as exercise type, intensity, duration, food type, and intake as input, and the extracted physiological response characteristics as output. The model is trained to obtain the physiological response coefficient corresponding to each program characteristic, and the quantitative physiological response value is calculated.

[0073] The evaluation module receives physiological response values. For example, if a particular exercise reduces a user's resting heart rate by a predetermined threshold, the exercise is considered to meet the physiological response criteria. The evaluation module then integrates the achievement of each physiological indicator to calculate an overall effectiveness score for the current program.

[0074] The adjustment module receives the overall effect score. If the score is below a preset value, it uses the model trained by the modeling module to predict the expected changes in key physiological indicators (such as weight and blood pressure) from different adjustment plans, such as increasing exercise duration or adjusting diet. The system selects the adjustment plan that can achieve the expected changes in key indicators and generates a new health management plan that is pushed to the user.

[0075] Furthermore, the modeling module includes:

[0076] Receiving module: Receives individual physiological indicator data and program execution information, and pre-processes the individual physiological indicator data, removes abnormal values ​​and missing values, and obtains valid physiological indicator data;

[0077] Extraction module: Use time series analysis method to analyze the dynamic change trend of effective physiological indicator data before and after the implementation of a specific plan, and extract the change amplitude, change rate and duration as physiological response characteristics;

[0078] Modeling submodule: takes the information of specific program execution as input and the physiological response characteristics as output, establishes the correlation model between the changes of individual physiological indicators and the specific program, and obtains the physiological response coefficient corresponding to the specific program;

[0079] Calculation module: Calculates the individual's physiological response to a specific plan based on the physiological response coefficient to obtain a quantitative physiological response value.

[0080] Specifically, in order to address the problem that the existing technology lacks specific methods for data processing, trend analysis, feature extraction and model building in the modeling process, resulting in insufficient accuracy of modeling results, this solution introduces a receiving module to preprocess the original physiological indicator data, ensuring the quality and reliability of the data and avoiding modeling deviations caused by data problems.

[0081] Among them, the extraction module adopts the time series analysis method, provides a specific means to analyze the dynamic change trend of physiological indicators, and defines quantifiable physiological response characteristics such as change amplitude, change rate and duration, solving the problem of lack of specific methods for feature extraction.

[0082] The modeling submodule provides a specific technical path for establishing the association model by establishing the association model between specific scheme information and physiological response characteristics and outputting the physiological response coefficient, thereby quantifying the impact of different schemes on individual physiological indicators.

[0083] Finally, the calculation module calculates the quantified physiological response value based on the physiological response coefficient. This directly reflects the physiological changes after the individual implements the program, providing accurate input for subsequent program evaluation. Thus, this program provides a complete and specific operational modeling process, improving the accuracy and reliability of quantifying individual physiological responses.

[0084] Furthermore, the information of specific program execution includes exercise type, exercise intensity, exercise duration, food intake, and food type; the modeling submodule includes:

[0085] The first modeling module: construct a linear regression model including L1 regularization term as the association model between changes in individual physiological indicators and specific plans;

[0086] The second modeling module: Based on historical health management data, the L1 regularization coefficient of the linear regression model is optimized to obtain the optimized linear regression model;

[0087] The third modeling module: One-hot coding is performed on the exercise type and food type. The coded exercise type, food type, exercise intensity, exercise duration, and food intake are input into the optimized linear regression model to calculate the physiological response coefficient corresponding to the specific plan.

[0088] Specifically, this technical solution aims to address the problem that existing health management programs have difficulty capturing differences in individual physiological responses to specific interventions, leading to uncertain program effectiveness and biased risk assessments.

[0089] First, the system obtains and preprocesses the user's individual physiological indicator data and program execution information through the receiving module to ensure the validity of the data.

[0090] Next, the extraction module uses time series analysis methods to extract dynamic change trend characteristics before and after the implementation of a specific plan from the effective physiological indicator data. These characteristics (change amplitude, change rate, and duration) objectively quantify the immediate or short-term response of the physiological indicators to the plan.

[0091] Then, the modeling submodule receives these physiological response characteristics and the corresponding specific program execution information (including exercise type, exercise intensity, exercise duration, food intake, and food type).

[0092] To model the association between changes in individual physiological indicators and specific regimens, the first modeling module constructed a linear regression model with an L1 regularization term. This linear regression model establishes a linear relationship between regimen factors and physiological response characteristics. The L1 regularization term, by constraining the model coefficients, enables the model to automatically select key regimen factors that have a significant impact on physiological responses and compresses the coefficients of unimportant factors to zero. This enables feature selection, improves the model's generalization ability, avoids overfitting, and enables the model to more accurately identify and quantify the impact of different regimen components on individual physiological responses.

[0093] The second modeling module optimizes the L1 regularization coefficient of the linear regression model using historical health management data. Using techniques such as cross-validation, the regularization strength is adjusted based on historical data to find an optimal regularization coefficient. This balances the model's performance between fitting known data and predicting unknown data, further improving its accuracy and robustness.

[0094] Before inputting the specific plan information into the optimized model, the third modeling module performs one-hot encoding on the categorical features (exercise type, food type) and converts them into numerical vectors, avoiding errors or biases that may be introduced by directly using the categorical data. Then, the encoded categorical features are integrated with the numerical features (exercise intensity, exercise duration, food intake) to form a complete feature vector, which is input into the optimized linear regression model. Based on the learned association relationship, the model calculates and outputs the physiological response coefficients corresponding to the specific plan. These coefficients quantify the expected impact of the specific plan on the changes in individual physiological indicators. Finally, the calculation module calculates a quantitative physiological response value based on these physiological response coefficients, which comprehensively reflects the individual's physiological response to the specific plan. Through this series of steps, the technical solution can more effectively learn the individual physiological response patterns from multi-dimensional plan information, identify key influencing factors, and thus establish a more accurate association model, calculate physiological response coefficients and physiological response values ​​that better reflect individual differences, solve the problem that existing methods are difficult to accurately quantify individual physiological responses, and provide a quantitative basis for subsequent plan evaluation and adjustment.

[0095] Furthermore, the third modeling module includes:

[0096] Encoding module: One-hot encodes the sports type and food type to obtain the corresponding vector code;

[0097] Integration module: Combines the vector code with the values ​​of exercise intensity, exercise duration, and food intake to obtain the feature vector of the specific plan;

[0098] The first calculation module: input the characteristic vector into the optimized linear regression model to calculate the physiological response coefficient corresponding to the specific solution.

[0099] Specifically, in order to solve the problem of integrating categorical data and numerical data as model input, this solution introduces an encoding module to preprocess the categorical data.

[0100] The encoding module receives the exercise type and food type information and converts it into a numerical vector representation using the One-Hot encoding method. For example, if the exercise types include "running", "swimming", and "walking", "running" may be encoded as [1, 0, 0], while "swimming" is encoded as [0, 1, 0]. Similarly, the food type is also converted into a vector code. Subsequently, the integration module receives these vector codes as well as numerical data such as exercise intensity, exercise duration, and food intake. The integration module splices the numerical representations from these different sources to form a single feature vector. This feature vector contains all the relevant input information describing a specific health management plan and is presented in a unified format required by the linear regression model. Finally, the first calculation module receives the feature vector output by the integration module and inputs it into the pre-optimized linear regression model.

[0101] The linear regression model uses this eigenvector to calculate and output the physiological response coefficient corresponding to the specific regimen. This step-by-step process ensures the standardization of the input data format, allowing the linear regression model to accurately calculate the physiological response coefficient, thereby quantifying the individual's physiological response to the specific regimen.

[0102] Furthermore, the first calculation module includes:

[0103] The second calculation module: inputs the feature vector into the optimized linear regression model and outputs the physiological response prediction value;

[0104] The third calculation module: uses the formula: physiological impact coefficient = (physiological impact prediction value * standard deviation of physiological indicator data) + mean value of physiological indicator data to calculate the physiological impact coefficient corresponding to the specific plan.

[0105] Specifically, the first calculation module is used to calculate the physiological impact coefficient corresponding to a specific plan. First, the second calculation module receives the feature vector of the specific plan after encoding and integration. This feature vector contains information such as exercise type, exercise intensity, exercise duration, food intake, and food type.

[0106] The second computational module feeds this feature vector into a linear regression model that has been pre-trained and optimized based on historical health management data. Based on the input features, the model outputs a physiological response prediction value, which is the model's initial estimate of the physiological changes that the proposed solution may cause.

[0107] Next, the third calculation module receives the physiological response prediction value and converts it into a meaningful physiological impact coefficient using a formula: the physiological impact coefficient is equal to the physiological impact prediction value multiplied by the standard deviation of the physiological indicator data, plus the mean of the physiological indicator data.

[0108] Through this transformation, the predicted values ​​output by the linear regression model, which may be normalized, are mapped back to the scale of the original physiological indicator data, resulting in a quantitative physiological impact coefficient. This coefficient intuitively represents the expected impact of a specific program on the physiological indicators of a specific individual.

[0109] Furthermore, the assessment modules include:

[0110] Acquisition module: Acquisition of a preset health management program evaluation index system, which includes multiple physiological index threshold ranges related to sub-health and corresponding weight coefficients;

[0111] The fourth calculation module: compares the physiological response value with the physiological indicator threshold range in the evaluation indicator system, determines the compliance status of each physiological indicator, and calculates the comprehensive compliance rate based on the corresponding weight coefficient;

[0112] The fifth calculation module: Based on the comprehensive compliance rate, program implementation compliance and user feedback score, calculate the overall effect score of the current health management program on the user and obtain a quantitative evaluation result.

[0113] Please refer to Figure 3 Specifically, when evaluating the overall effect of a health management program on a user, the acquisition module first obtains a preset health management program evaluation index system. This evaluation index system is stored in a database and includes health threshold ranges set for different sub-health-related physiological indicators (such as blood pressure, blood sugar, weight, etc.) and the weight of each indicator in the overall evaluation.

[0114] Subsequently, the fourth calculation module receives the physiological response value output by the modeling module. The physiological response value reflects the degree of change in the user's physiological indicators relative to the baseline or target value after the user implements a specific health management plan. The fourth calculation module compares these physiological response values ​​with the corresponding physiological indicator threshold range in the evaluation index system. For example, if the physiological response value of a physiological indicator falls within the health threshold range, the indicator is judged to be up to standard. Based on the compliance status of each physiological indicator and its preset weight in the evaluation index system, the fourth calculation module calculates the comprehensive compliance rate. The comprehensive compliance rate quantifies the effect of the program in improving the user's physiological health status.

[0115] The fifth calculation module integrates plan compliance data and user feedback scores based on the comprehensive compliance rate. Plan compliance reflects the degree to which users actually implement the recommended plan, and is calculated, for example, by recording user check-ins and task completion. User feedback scores are collected through questionnaires and other methods, reflecting users' subjective feelings and satisfaction with the plan.

[0116] The fifth calculation module uses a weighted average method to combine the comprehensive target achievement rate, program compliance, and user feedback scores to produce a quantitative overall effectiveness score. This score comprehensively reflects the effectiveness of the program across multiple dimensions, including physiological improvement, user behavior, and subjective experience. It overcomes the limitations of relying solely on physiological response values ​​for evaluation and provides a more comprehensive and accurate basis for subsequent program adjustments.

[0117] Furthermore, the fifth calculation module includes:

[0118] The sixth calculation module: obtains the user's plan execution records during the health management plan execution cycle, and calculates the plan execution compliance score based on the degree of match between the plan items actually executed and the recommended plan items in the records, where the plan execution compliance score = (the number of plan items actually executed / the total number of recommended plan items) * 100%;

[0119] The seventh calculation module collects feedback questionnaires filled out by users after the health management plan execution cycle. The questionnaires cover multiple dimensions such as satisfaction with the plan, ease of use, and perceived effectiveness. The questionnaire results are normalized to obtain user feedback scores.

[0120] The eighth calculation module: Using the weighted average method, the comprehensive compliance rate, program implementation compliance score and user feedback score are weighted and calculated to obtain the overall effect score of the current health management program on the user.

[0121] The sixth calculation module accesses the user's activity log stored in the system or the execution record manually entered by the user to identify the health management task items actually completed by the user. These task items are compared with the solution items recommended by the system to calculate the number of matches.

[0122] The compliance score was calculated as a simple ratio of the number of items actually completed to the total number of recommended items.

[0123] The seventh calculation module collects subjective feedback by pushing a structured questionnaire to users. The questionnaire design contains multiple evaluation dimensions that are quantitative or can be converted into quantitative scores.

[0124] Normalization of questionnaire results can unify feedback of different dimensions or scoring ranges into a standardized numerical range, such as 0 to 1 or 0 to 100.

[0125] The eighth calculation module receives the comprehensive compliance rate, compliance score and normalized user feedback score as input, and performs a weighted summation operation based on a preset weight coefficient to obtain the final overall effect score.

[0126] Furthermore, the adjustment module includes:

[0127] Prediction module: If the quantitative assessment result is lower than the preset effect threshold, the module predicts the expected change in the user's physiological indicator after adjusting at least one of the following: exercise type, exercise intensity, exercise duration, food intake, and food type, based on the correlation model between individual physiological indicator changes and specific plans;

[0128] Adjustment submodule: Select an adjustment plan that makes the expected change value of at least one key physiological indicator reach the preset target, and combine it with the preset plan adjustment rule library to make constrained adjustments to the user's health management plan, and generate the adjusted health management plan and push it to the user.

[0129] Please refer to Figure 4 ,Specifically, this technical solution describes the specific implementation of the ,adjustment module, aiming to solve the problem that the existing ,adjustment mechanism lacks predictive capabilities leading to uncertain ,adjustment effects.

[0130] When the quantitative evaluation results of the current health management plan do not meet the preset standards, the system starts the adjustment process. The prediction module intervenes first, using the previously established association model between changes in individual physiological indicators and specific plans. The association model reflects the physiological response patterns of specific users to different exercise and dietary interventions. Based on this association model, the prediction module simulates the expected changes in the user's key physiological indicators (such as blood pressure, blood sugar, weight, etc.) after making various possible adjustments to the current plan (for example, increasing exercise duration, changing exercise type, adjusting food intake or type).

[0131] The system then obtains data on the potential for improvement in physiological indicators corresponding to different adjustment plans. The adjustment submodule then receives this predicted data. The adjustment submodule sets specific physiological indicator improvement targets, such as reducing the systolic blood pressure of a hypertensive patient to below a specific value.

[0132] The adjustment submodule selects, from among the multiple adjustment options provided by the prediction module, a plan that achieves or exceeds the expected change in at least one key physiological indicator. After selecting a preliminary adjustment plan, the adjustment submodule applies constraints based on a pre-defined rule base. These rule bases contain constraints to ensure the safety and feasibility of the plan, such as limiting exercise intensity to a specific percentage of the user's maximum heart rate or limiting total daily food intake to a specific percentage of the user's total daily energy requirement.

[0133] These constraints ensure that the adjusted plan not only effectively improves physiological indicators but also meets the user's physiological tolerance and health principles. Ultimately, the system generates an adjusted health management plan, which has undergone prediction and constraint screening, and pushes it to the user for execution. By introducing a model-based prediction and goal-based constraint selection mechanism, this technical solution makes the health management plan adjustment process more intelligent and personalized, increasing the effectiveness of the adjustment plan in improving the user's health status, thereby more accurately managing the insurance risks of sub-healthy people.

[0134] Furthermore, in the adjustment submodule, the constraint adjustment includes: ensuring that the adjusted exercise intensity does not exceed 80% of the user's maximum heart rate, and ensuring that the adjusted food intake does not exceed 120% of the user's total daily energy requirement.

[0135] Among them, when the adjustment submodule generates the adjusted health management plan, it calculates the exercise intensity and food intake involved in the plan. The calculation of exercise intensity can estimate the corresponding target heart rate based on the recommended exercise type, duration and user individual data (such as age, gender, weight). The user's maximum heart rate can be estimated by a standard formula (such as 220 minus age). The system compares the estimated target heart rate with 80% of the user's maximum heart rate. The calculation of food intake can estimate its total energy based on the recommended food type and intake. The total energy required by the user per day can be estimated by the basal metabolic rate (BMR) combined with the activity level. The system compares the estimated total energy with 120% of the total energy required by the user per day. Therefore, in the process of generating the plan, the system will check whether these calculation results meet the preset constraints.

[0136] Furthermore, the acquisition module includes:

[0137] Monitoring module: real-time monitoring and recording of the user's individual physiological indicators, including heart rate, blood pressure, blood sugar, weight, sleep duration, and activity level;

[0138] Encryption module: uploads individual physiological indicator data and program execution information to the cloud server through a data encryption channel. The program execution information includes exercise type, exercise intensity, exercise duration, food intake, and food type;

[0139] Storage module: Receives individual physiological indicator data and program execution information, and writes the data into the database in timestamp order to ensure the timeliness and consistency of the data.

[0140] Please refer to Figure 2 ,Specifically, this technical solution provides a specific implementation method for ,the accuracy, security or lack of necessary sequential ,information that may arise during the data collection process.

[0141] The monitoring module continuously collects physiological indicators such as heart rate, blood pressure, blood sugar, weight, sleep duration, and activity level in real time, ensuring timely and comprehensive data. This provides the foundation for subsequent analysis of dynamic changes in physiological indicators. The encryption module encrypts data via a data encryption channel, such as the TLS / SSL protocol, during data upload to the cloud server. This ensures the secure transmission of individual physiological indicator data and plan execution information, preventing unauthorized access or tampering during data transmission and protecting user privacy.

[0142] Upon receiving the data, the storage module immediately attaches a precise timestamp to each piece of data and writes the data to the database in timestamp order. Data temporality is crucial for analyzing the dynamic changes in physiological indicators before and after protocol execution, accurately capturing changing trends; data consistency ensures the reliability of data analysis. Through these technical means, the acquisition module is able to obtain accurate, secure, and well-structured individual physiological indicator data and protocol execution information, providing high-quality input for subsequent modules to process, model, evaluate, and adjust data.

[0143] In this document, relational terms such as first and second, etc. are used merely to distinguish one entity or operation from another entity or operation, but do not necessarily require or imply any actual relationship or order between these entities or operations.

[0144] The foregoing is merely an embodiment of the present application and is not intended to limit the scope of protection of the present application. Persons skilled in the art will readily appreciate that the present application may be modified and altered in various ways. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present application shall be included within the scope of protection of the present application.

Claims

1. A system for insurance risk assessment and health management plan generation for sub-healthy people, characterized by: The system includes: Collection module: collects individual physiological indicator data and plan execution information of users during the implementation of health management plans; Modeling module: receiving the individual physiological indicator data and the program execution information, analyzing the dynamic change trend of the individual physiological indicator data after the specific program is executed in the program execution information, and establishing a correlation model between the individual physiological indicator changes and the specific program to quantify the individual's physiological response to the specific program and obtain a physiological response value; Evaluation module: evaluates the overall effect of the current health management program on the user based on the physiological response value and obtains a quantitative evaluation result; Adjustment module: adjust the user's health management plan according to the quantitative evaluation results.

2. The insurance risk assessment and health management plan generation system for sub-healthy people according to claim 1 is characterized in that: The modeling modules include: Receiving module: receiving the individual physiological indicator data and the program execution information, and preprocessing the individual physiological indicator data to eliminate abnormal values ​​and missing values ​​to obtain valid physiological indicator data; Extraction module: using a time series analysis method to analyze the dynamic change trend of the effective physiological indicator data before and after the execution of the specific plan, and extracting the change amplitude, change rate and duration as physiological response characteristics; Modeling submodule: taking the information of the execution of the specific plan as input and the physiological response characteristics as output, establishing a correlation model between the changes in individual physiological indicators and the specific plan, and obtaining the physiological response coefficient corresponding to the specific plan; Calculation module: Calculates the individual's physiological response degree to the specific program based on the physiological response coefficient to obtain a quantified physiological response value.

3. The insurance risk assessment and health management plan generation system for sub-healthy people according to claim 2 is characterized in that: Information on specific program implementation includes exercise type, exercise intensity, exercise duration, food intake, and food type; The modeling submodules include: The first modeling module: construct a linear regression model including L1 regularization term as the association model between changes in individual physiological indicators and specific plans; The second modeling module: based on the historical health management data, optimizes the L1 regularization coefficient of the linear regression model to obtain an optimized linear regression model; The third modeling module: One-hot encoding is performed on the exercise type and food type, and the encoded exercise type, food type, exercise intensity, exercise duration, and food intake are input into the optimized linear regression model to calculate the physiological response coefficient corresponding to the specific plan.

4. The insurance risk assessment and health management plan generation system for sub-healthy people according to claim 3 is characterized in that: The third modeling module includes: Encoding module: performing one-hot encoding on the exercise type and the food type to obtain corresponding vector codes; Integration module: splicing and integrating the vector code with the values ​​of exercise intensity, exercise duration, and food intake to obtain a feature vector of a specific plan; The first calculation module is to input the characteristic vector into the optimized linear regression model to calculate the physiological response coefficient corresponding to the specific solution.

5. The insurance risk assessment and health management plan generation system for sub-healthy people according to claim 3 is characterized in that: The first calculation module includes: A second calculation module: inputs the feature vector into the optimized linear regression model and outputs a physiological response prediction value; The third calculation module: uses the formula: physiological impact coefficient = (physiological impact prediction value * standard deviation of physiological indicator data) + mean value of physiological indicator data to calculate the physiological impact coefficient corresponding to the specific plan.

6. The insurance risk assessment and health management plan generation system for sub-healthy people according to claim 1 is characterized in that: Assessment modules include: Acquisition module: Acquisition of a preset health management program evaluation index system, wherein the evaluation index system includes a plurality of physiological index threshold ranges related to sub-health and corresponding weight coefficients; A fourth calculation module: comparing the physiological response value with the physiological indicator threshold range in the evaluation indicator system, determining the compliance status of each physiological indicator, and calculating the comprehensive compliance rate according to the corresponding weight coefficient; The fifth calculation module: based on the comprehensive compliance rate, program implementation compliance and user feedback score, calculate the overall effect score of the current health management program on the user to obtain a quantitative evaluation result.

7. The insurance risk assessment and health management plan generation system for sub-healthy people according to claim 6 is characterized in that: The fifth calculation module includes: The sixth calculation module: obtains the user's plan execution records during the health management plan execution cycle, and calculates the plan execution compliance score based on the degree of match between the plan items actually executed and the recommended plan items in the records, where the plan execution compliance score = (the number of plan items actually executed / the total number of recommended plan items) * 100%; The seventh calculation module collects feedback questionnaires filled out by users after the health management plan execution cycle. The questionnaires cover multiple dimensions such as satisfaction with the plan, ease of use, and perceived effectiveness. The questionnaire results are normalized to obtain user feedback scores. The eighth calculation module: Using the weighted average method, the comprehensive compliance rate, program implementation compliance score and user feedback score are weighted and calculated to obtain the overall effect score of the current health management program on the user.

8. The insurance risk assessment and health management plan generation system for sub-healthy people according to claim 1 is characterized in that: The adjustment module includes: Prediction module: If the quantitative assessment result is lower than a preset effect threshold, the prediction module predicts the expected change in the user's physiological indicator after adjusting at least one of the exercise type, exercise intensity, exercise duration, food intake, and food type based on the correlation model between the change in individual physiological indicators and the specific plan; Adjustment submodule: Select an adjustment plan that makes the expected change value of at least one key physiological indicator reach the preset target, and combine it with the preset plan adjustment rule library to make constrained adjustments to the user's health management plan, and generate the adjusted health management plan and push it to the user.

9. The insurance risk assessment and health management plan generation system for sub-healthy people according to claim 8, characterized in that: In the adjustment submodule, the constraint adjustment includes: ensuring that the adjusted exercise intensity does not exceed 80% of the user's maximum heart rate, and ensuring that the adjusted food intake does not exceed 120% of the user's total daily energy requirement.

10. The insurance risk assessment and health management plan generation system for sub-healthy people according to claim 1, characterized in that: The acquisition module includes: Monitoring module: real-time monitoring and recording of the user's individual physiological indicators, including heart rate, blood pressure, blood sugar, weight, sleep duration, and activity level; Encryption module: uploads individual physiological indicator data and program execution information to the cloud server through a data encryption channel. The program execution information includes exercise type, exercise intensity, exercise duration, food intake, and food type; Storage module: receives the individual physiological indicator data and the program execution information, and writes the data into the database in the order of timestamp to ensure the time sequence and consistency of the data.

Images