Multi-parameter fusion cardiovascular disease risk quantitative evaluation system
The cardiovascular disease risk quantification assessment system, which integrates multiple parameters, utilizes a deep neural network model to scientifically allocate weights. This solves the problems of existing assessment models relying on a few parameters and being unable to dynamically assess risks, enabling precise quantification and personalized assessment of cardiovascular disease risk and improving the accuracy and convenience of the assessment.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-05
- Publication Date
- 2026-03-13
AI Technical Summary
Existing cardiovascular disease risk assessment models rely on a few key parameters, lack precise quantification, cannot perform real-time dynamic assessments, and cannot effectively handle the correlations and interactions between multiple parameters, resulting in insufficient accuracy of assessment results.
A multi-parameter fusion cardiovascular disease risk quantification assessment system is adopted. Through data acquisition, preprocessing, multi-parameter fusion assessment model and dynamic tracking module, combined with deep neural network model, weights are scientifically allocated to achieve accurate quantification and personalized assessment of cardiovascular disease risk.
It improves the accuracy and comprehensiveness of assessments, provides intuitive quantitative indicators, supports dynamic monitoring and early intervention, and seamlessly integrates with wearable devices to ensure data security and privacy.
Smart Images
Figure CN121662383A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of medical and health technology, specifically to a multi-parameter fusion system for quantitative assessment of cardiovascular disease risk. Background Technology
[0002] Cardiovascular disease is one of the leading causes of death worldwide, and its development involves the combined effects of multiple risk factors. Currently, widely used international cardiovascular disease risk assessment models include the Framingham risk assessment model, the European SCORE risk assessment model, the WHO / ISH risk prediction map, and the Chinese ischemic cardiovascular disease risk assessment model. However, these models have the following limitations: Existing models mainly rely on a few key parameters (such as age, sex, blood pressure, cholesterol levels, etc.) and fail to fully consider the combined effects of multiple physiological parameters. The assessment results are mostly qualitative or semi-quantitative descriptions, lacking precise quantitative indicators, making it difficult to conduct individualized risk classification and intervention; It is impossible to assess individual risk changes in real time and to track risk changes over different time periods; For younger populations, existing models often underestimate their future cardiovascular disease risk and fail to accurately reflect changes in relative risk. The assessment process relies on the doctor's professional judgment and lacks standardized and automated procedures.
[0003] In recent years, with the popularization of wearable devices and the accumulation of medical big data, multi-parameter fusion analysis has become a new trend in cardiovascular disease risk assessment. However, existing multi-parameter fusion methods often lack a scientific weight allocation mechanism, failing to effectively handle the correlation and mutual influence between different parameters, resulting in insufficient accuracy of assessment results. Summary of the Invention
[0004] (a) Technical problems to be solved To address the shortcomings of existing technologies, this invention provides a multi-parameter fusion-based cardiovascular disease risk quantification assessment system. By integrating various physiological parameters, lifestyle habits, genetic factors, and environmental factors, a scientific multi-parameter fusion assessment model is established to achieve accurate quantitative assessment of cardiovascular disease risk.
[0005] (II) Technical Solution To achieve the above objectives, the present invention provides the following technical solution: a multi-parameter fusion cardiovascular disease risk quantification assessment system, comprising a data acquisition module, a data preprocessing module, a multi-parameter fusion assessment model, a risk quantification output module, and a dynamic tracking module, wherein the multi-parameter fusion assessment model comprises a feature selection submodule, a weight allocation submodule, a risk assessment submodule, and a risk grading submodule; The feature selection submodule filters out the features with the highest correlation to cardiovascular disease risk from the preprocessed data. The weight allocation submodule assigns scientific weights to different features based on medical knowledge and historical data. The risk assessment submodule uses a deep neural network model to map multi-parameter inputs to cardiovascular disease risk quantification values. The risk grading submodule divides users into three levels: low risk, medium risk, and high risk, based on the risk quantification values. The dynamic tracking module collects user data periodically, establishes a trend chart of user risk changes, and promptly detects risk fluctuations. When the risk value exceeds a preset threshold, the system automatically sends an early warning notification and suggests that the user undergo further medical examinations.
[0006] Preferably, the data acquisition module includes a basic physiological parameter acquisition unit, a lifestyle parameter acquisition unit, a genetic information acquisition unit, and an environmental factor acquisition unit. The basic physiological parameter acquisition unit collects the user's heart rate, blood pressure, blood lipids, blood sugar, and other physiological parameters in real time through a smart bracelet or a smart blood pressure monitor wearable device. The lifestyle parameter collection unit collects user data on diet, exercise, and sleep habits via a mobile application. The genetic information collection unit obtains genetic risk information through gene testing services or family medical history questionnaires; The environmental factor acquisition unit obtains air quality, temperature, and humidity information of the user's environment through GPS positioning and meteorological API.
[0007] Preferably, the data preprocessing module includes a data cleaning unit, a data standardization unit, and a feature engineering unit. The data cleaning unit processes missing and outlier values by setting thresholds and using statistical methods. The data standardization unit converts data of different dimensions into data of a unified dimension. The feature engineering unit extracts key features.
[0008] Preferably, the risk quantification output module converts the assessment results into a quantitative risk value of 0-100 and provides a detailed risk factor analysis report, which includes an overall risk score, the contribution of each risk factor, a risk change trend chart, and targeted improvement suggestions.
[0009] Preferably, the construction process of the multi-parameter fusion evaluation model specifically includes the following steps: S1. Construct a cardiovascular disease risk assessment database, containing a large number of samples of known cardiovascular disease risk factors and eventual incidence. S2. Perform feature engineering on the database to extract key features; S3. A deep neural network model is used to learn the mapping relationship between different features and cardiovascular disease risk through training set data; S4. Optimize model parameters through cross-validation to ensure the model's generalization ability; S5. Based on medical knowledge and historical data, assign scientific weights to different characteristics.
[0010] Preferably, in the multi-parameter fusion evaluation model, data cleaning and missing value handling employ a statistical outlier detection method. This involves cleaning the original data, calculating the mean and standard deviation of each parameter, and considering values exceeding ±3σ as outliers. For missing values, multiple imputation is used. The specific steps are as follows: T1. Construct a missing value pattern matrix to identify patterns of missing data; T2. Construct a prediction model based on the K-nearest neighbor algorithm; T3. Generate multiple complete datasets using the Bootstrap method; T4. Perform parameter estimation for each complete dataset; T5. The final missing value filling is obtained by combining multiple estimation results.
[0011] Preferably, the extraction of key features in step S2 specifically includes: Basic physiological characteristics: heart rate variability, blood pressure variability, and blood glucose variability; Time series characteristics: autocorrelation coefficient, spectral characteristics, and trend components; Multimodal characteristics: cross-analysis of physiological parameters with lifestyle habits and environmental factors.
[0012] Preferably, this multi-parameter fusion cardiovascular disease risk quantification assessment system is seamlessly integrated with wearable devices and mobile medical applications to achieve convenient daily monitoring. Moreover, the system adopts privacy protection technology to ensure the security and privacy of user data. The system can provide targeted risk assessment and improvement suggestions based on the user's age, gender, health status, and other personalized factors.
[0013] (III) Beneficial Effects This invention provides a multi-parameter fusion-based system for quantitative assessment of cardiovascular disease risk. Compared with existing technologies, it has the following advantages: (1) The multi-parameter fusion cardiovascular disease risk quantification assessment system comprehensively considers the multi-dimensional factors of cardiovascular disease risk through multi-parameter fusion, which improves the accuracy and comprehensiveness of the assessment. The deep learning-based assessment model can automatically learn the complex relationship between parameters, avoiding the limitations of manually setting weights.
[0014] (2) The multi-parameter fusion cardiovascular disease risk quantification assessment system has intuitive and easy-to-understand risk quantification indicators, which are easy for users to understand and accept. The dynamic tracking function realizes continuous monitoring of user risk, which helps early intervention. The system can be seamlessly integrated with wearable devices and mobile medical applications to achieve convenient daily monitoring. Attached Figure Description
[0015] Figure 1 This is a block diagram illustrating the structural principle of the system of the present invention. Detailed Implementation
[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0017] Please see Figure 1 The present invention provides two technical solutions: a multi-parameter fusion cardiovascular disease risk quantification assessment system, specifically including the following embodiments: Example 1: A multi-parameter fusion cardiovascular disease risk quantification assessment system, including a data acquisition module, a data preprocessing module, a multi-parameter fusion assessment model, a risk quantification output module, and a dynamic tracking module. The multi-parameter fusion assessment model includes a feature selection submodule, a weight allocation submodule, a risk assessment submodule, and a risk grading submodule. The feature selection submodule filters out the features with the highest correlation to cardiovascular disease risk from the preprocessed data. The weight allocation submodule assigns scientific weights to different features based on medical knowledge and historical data. The risk assessment submodule uses a deep neural network model to map multi-parameter inputs to a cardiovascular disease risk quantification value. The risk grading submodule divides users into three levels: low risk, medium risk, and high risk, based on the risk quantification value. The dynamic tracking module collects user data regularly to create a trend chart of user risk changes, promptly detects risk fluctuations, and automatically sends an early warning notification when the risk value exceeds a preset threshold, and recommends that the user undergo further medical examination.
[0018] In this embodiment of the invention, the data acquisition module includes a basic physiological parameter acquisition unit, a lifestyle parameter acquisition unit, a genetic information acquisition unit, and an environmental factor acquisition unit. The basic physiological parameter acquisition unit collects the user's heart rate, blood pressure, blood lipids, blood sugar, and other physiological parameters in real time through a smart bracelet and a smart blood pressure monitor wearable device. The lifestyle habit parameter collection unit collects users' diet, exercise, and sleep habit data through a mobile application; The genetic information collection unit obtains genetic risk information through gene testing services or family medical history questionnaires; The environmental factor acquisition unit obtains air quality, temperature, and humidity information of the user's environment through GPS positioning and meteorological API.
[0019] In this embodiment of the invention, the data preprocessing module includes a data cleaning unit, a data standardization unit, and a feature engineering unit. The data cleaning unit processes missing and outlier values by setting thresholds and using statistical methods. The data standardization unit converts data of different dimensions into a unified dimension, such as using the Z-score standardization method. The feature engineering unit extracts key features, such as heart rate variability (HRV), blood pressure variability (BPV), and blood glucose variability.
[0020] In this embodiment of the invention, the risk quantification output module converts the evaluation results into a quantitative risk value of 0-100 and provides a detailed risk factor analysis report, which includes an overall risk score, the contribution of each risk factor, a risk change trend chart, and targeted improvement suggestions.
[0021] Example 2: The technical solution of this embodiment of the invention differs from that of Example 1 in that the construction process of the multi-parameter fusion evaluation model specifically includes the following steps: S1. Construct a cardiovascular disease risk assessment database, containing a large number of samples of known cardiovascular disease risk factors and eventual incidence. S2. Perform feature engineering on the database to extract key features; S3. A deep neural network model is used to learn the mapping relationship between different features and cardiovascular disease risk through training set data; S4. Optimize model parameters through cross-validation to ensure the model's generalization ability; S5. Based on medical knowledge and historical data, assign scientific weights to different characteristics.
[0022] In this embodiment of the invention, the data cleaning and missing value handling in the multi-parameter fusion evaluation model employs a statistical outlier detection method. The original data is cleaned, and the mean and standard deviation of each parameter are calculated. Values exceeding ±3σ are considered outliers. For missing values, multiple imputation is used. The specific steps are as follows: T1. Construct a missing value pattern matrix to identify patterns of missing data; T2. Construct a prediction model based on the K-nearest neighbor algorithm; T3. Generate multiple complete datasets using the Bootstrap method; T4. Perform parameter estimation for each complete dataset; T5. The final missing value filling is obtained by combining multiple estimation results.
[0023] In this embodiment of the invention, the extraction of key features in step S2 specifically includes: Basic physiological characteristics: heart rate variability, blood pressure variability, and blood glucose variability; Time series characteristics: autocorrelation coefficient, spectral characteristics, and trend components; Multimodal characteristics: cross-analysis of physiological parameters with lifestyle habits and environmental factors.
[0024] In this embodiment of the invention, the multi-parameter fusion cardiovascular disease risk quantification assessment system is seamlessly integrated with wearable devices and mobile medical applications to achieve convenient daily monitoring. Moreover, the system adopts privacy protection technology to ensure the security and privacy of user data. The system can provide targeted risk assessment and improvement suggestions based on the user's age, gender, health status, and other personalized factors.
[0025] Multi-parameter fusion evaluation model construction: Deep neural networks (DNNs) are used as the core architecture of the fusion model, which includes the following layers: Input layer: Receives preprocessed multidimensional feature vectors Hidden layers: Contain 3 fully connected layers, with 128, 64, and 32 nodes in each layer respectively. Dropout layer: Add a Dropout layer between hidden layers to prevent overfitting. Output layer: A single neuron outputs a quantitative value for cardiovascular disease risk.
[0026] Furthermore, any content not described in detail in this specification is existing technology known to those skilled in the art.
[0027] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.
[0028] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.
Claims
1. A multi-parameter fusion system for quantitative assessment of cardiovascular disease risk, characterized in that: It includes a data acquisition module, a data preprocessing module, a multi-parameter fusion evaluation model, a risk quantification output module, and a dynamic tracking module. The multi-parameter fusion evaluation model includes a feature selection submodule, a weight allocation submodule, a risk assessment submodule, and a risk classification submodule. The feature selection submodule filters out the features with the highest correlation to cardiovascular disease risk from the preprocessed data. The weight allocation submodule assigns scientific weights to different features based on medical knowledge and historical data. The risk assessment submodule uses a deep neural network model to map multi-parameter inputs to cardiovascular disease risk quantification values. The risk grading submodule divides users into three levels: low risk, medium risk, and high risk, based on the risk quantification values. The dynamic tracking module collects user data periodically, establishes a trend chart of user risk changes, and promptly detects risk fluctuations. When the risk value exceeds a preset threshold, the system automatically sends an early warning notification and suggests that the user undergo further medical examinations.
2. The multi-parameter fusion cardiovascular disease risk quantification assessment system according to claim 1, characterized in that: The data acquisition module includes a basic physiological parameter acquisition unit, a lifestyle parameter acquisition unit, a genetic information acquisition unit, and an environmental factor acquisition unit. The basic physiological parameter acquisition unit collects the user's heart rate, blood pressure, blood lipids, blood sugar, and other physiological parameters in real time through smart bracelets and smart blood pressure monitors. The lifestyle parameter collection unit collects user data on diet, exercise, and sleep habits via a mobile application. The genetic information collection unit obtains genetic risk information through gene testing services or family medical history questionnaires; The environmental factor acquisition unit obtains air quality, temperature, and humidity information of the user's environment through GPS positioning and meteorological API.
3. The multi-parameter fusion cardiovascular disease risk quantification assessment system according to claim 1, characterized in that: The data preprocessing module includes a data cleaning unit, a data standardization unit, and a feature engineering unit. The data cleaning unit processes missing and outlier values by setting thresholds and using statistical methods. The data standardization unit converts data of different dimensions into data of a unified dimension. The feature engineering unit extracts key features.
4. The multi-parameter fusion cardiovascular disease risk quantification assessment system according to claim 1, characterized in that: The risk quantification output module converts the assessment results into a quantitative risk value of 0-100 and provides a detailed risk factor analysis report, which includes an overall risk score, the contribution of each risk factor, a risk change trend chart, and targeted improvement suggestions.
5. The multi-parameter fusion cardiovascular disease risk quantification assessment system according to claim 1, characterized in that: The construction process of the multi-parameter fusion evaluation model specifically includes the following steps: S1. Construct a cardiovascular disease risk assessment database, containing a large number of samples of known cardiovascular disease risk factors and eventual incidence. S2. Perform feature engineering on the database and extract key features; S3. A deep neural network model is used to learn the mapping relationship between different features and cardiovascular disease risk through training set data; S4. Optimize model parameters through cross-validation to ensure the model's generalization ability; S5. Based on medical knowledge and historical data, assign scientific weights to different characteristics.
6. The multi-parameter fusion cardiovascular disease risk quantification assessment system according to claim 5, characterized in that: The data cleaning and missing value handling in the multi-parameter fusion evaluation model employs a statistical outlier detection method. The original data is cleaned, and the mean and standard deviation of each parameter are calculated. Values exceeding ±3σ are considered outliers. For missing values, multiple imputation is used. The specific steps are as follows: T1. Construct a missing value pattern matrix to identify patterns of missing data; T2. Construct a prediction model based on the K-nearest neighbor algorithm; T3. Generate multiple complete datasets using the Bootstrap method; T4. Perform parameter estimation for each complete dataset; T5. The final missing value filling is obtained by combining multiple estimation results.
7. The multi-parameter fusion cardiovascular disease risk quantification assessment system according to claim 5, characterized in that: The specific steps for extracting key features in step S2 include: Basic physiological characteristics: heart rate variability, blood pressure variability, and blood glucose variability; Time series characteristics: autocorrelation coefficient, spectral characteristics, and trend components; Multimodal characteristics: cross-analysis of physiological parameters with lifestyle habits and environmental factors.
8. The multi-parameter fusion cardiovascular disease risk quantification assessment system according to claim 1, characterized in that: This multi-parameter fusion cardiovascular disease risk quantification assessment system is seamlessly integrated with wearable devices and mobile medical applications to achieve convenient daily monitoring. The system also employs privacy protection technology to ensure the security and privacy of user data. The system can provide targeted risk assessments and improvement suggestions based on personalized factors such as user age, gender, and health status.