Artificial Intelligence-Based Health Monitoring and Early Warning System for Chronic Diseases Related to Traditional Chinese Medicine Constitution

By using an AI-based TCM constitution-based chronic disease health monitoring and early warning system, which combines physiological, behavioral, and TCM physical signs information, the system addresses the shortcomings of existing technologies in analyzing constitution differences. It enables intelligent identification and dynamic early warning of chronic disease risks, improving the accuracy and personalization of early warnings.

CN123120375APending Publication Date: 2026-11-17BEIJING UNIV OF CHINESE MEDICINE
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611335581.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-31
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing chronic disease health monitoring systems lack the ability to analyze physical differences and make intelligent predictions, making it difficult to achieve personalized and dynamic health risk warnings. Furthermore, they are insufficient in the integration of multi-source health data and trend prediction.

Method used

An AI-based health monitoring and early warning system for chronic diseases related to TCM constitution is adopted, which includes modules for data collection, processing, construction of monitoring features, risk identification, and early warning determination. It combines physiological, behavioral, medication, and TCM physical signs information and uses an AI model to identify risks and determine early warnings.

Benefits of technology

It enables intelligent identification and dynamic early warning of chronic disease risks, improving the accuracy and personalization of early warnings, and comprehensively considering the health status and management implementation of chronic disease patients.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123120375A_ABST
    Figure CN123120375A_ABST
Patent Text Reader

Abstract

This invention relates to the field of health supervision and early warning technology, and discloses an artificial intelligence-based health supervision and early warning system for chronic diseases based on traditional Chinese medicine (TCM) constitution. The system includes a health data and TCM vital signs acquisition module, a data processing module, a supervision feature construction module, a risk identification module, and an early warning determination module. The acquisition module is used to obtain chronic disease-related data of chronic disease subjects, which includes at least physiological monitoring data and TCM vital signs information. The data processing module is used to perform time alignment and standardization processing on the chronic disease-related data to obtain basic supervision data. The supervision feature construction module is used to extract current state features, stage change features, and management features, and extract constitution identification-related features based on TCM vital signs information. The risk identification module is used to input the above features into an artificial intelligence model and output health risk identification results and TCM constitution classification results.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the fields of health management and smart healthcare, and in particular to an artificial intelligence-based health monitoring and early warning system for chronic diseases caused by traditional Chinese medicine constitution. Background Technology

[0002] Chronic disease management is typically characterized by long management cycles, continuous changes in condition, diverse influencing factors, and significant individual differences. In the process of chronic disease health monitoring, it is necessary not only to pay attention to physiological monitoring data such as blood pressure, blood sugar, and heart rate, but also to combine behavioral patterns, medication adherence, and past health information to continuously assess the health status of individuals with chronic diseases and provide risk warnings.

[0003] Existing chronic disease health-related technologies can collect, analyze, and provide alerts for some health data, but in the context of chronic disease health supervision and early warning, the following problems still exist: chronic disease-related data from different sources need to be further integrated to form basic data suitable for continuous supervision; judging based solely on single data or current data is insufficient to fully reflect the phased changes in the status of chronic disease subjects; in the process of chronic disease health supervision, the implementation of management has a real impact on risk assessment, and therefore it is necessary to incorporate it into the early warning assessment process. Existing health monitoring and chronic disease management systems mostly rely on single physiological indicators for threshold judgment, lacking the ability to comprehensively analyze differences in traditional Chinese medicine constitutions. At the same time, they are not intelligent enough in terms of multi-source health data fusion and trend prediction, making it difficult to achieve personalized and dynamic health risk warnings.

[0004] To address this, an artificial intelligence-based health monitoring and early warning system for chronic diseases related to TCM constitution is proposed. Summary of the Invention

[0005] In response to the above situation, the present invention provides an artificial intelligence-based TCM constitution-based chronic disease health monitoring and early warning system to solve the problem of the lack of constitution difference analysis and intelligent prediction capabilities in the existing technology, thereby realizing intelligent identification and dynamic early warning of chronic disease risks.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a health monitoring and early warning system for chronic diseases based on traditional Chinese medicine constitution using artificial intelligence, comprising a data acquisition module, a data processing module, a monitoring feature construction module, a risk identification module, and an early warning determination module.

[0007] The data acquisition module is used to collect chronic disease-related data of chronic disease subjects, and the chronic disease-related data includes at least physiological monitoring data.

[0008] The data processing module is connected to the data acquisition module and is used to perform time alignment and standardization processing on the chronic disease-related data to obtain basic regulatory data arranged in time series.

[0009] The regulatory feature construction module is connected to the data processing module and is used to extract current status features reflecting the health status of chronic disease subjects within the current time period based on the regulatory basic data. The current status features are then compared with the corresponding status data within historical time periods to form stage change features representing the stage changes in health status. Simultaneously, management features representing the implementation of chronic disease management are formed based on the management execution data of chronic disease subjects. Furthermore, relevant features for constitution identification are constructed by combining TCM physical signs information to build a regulatory feature dataset.

[0010] The risk identification module is connected to the regulatory feature construction module and is used to input the current status features, stage change features, management features and constitution identification-related features into the artificial intelligence model to obtain the health risk identification results of chronic disease subjects and output the TCM constitution classification results.

[0011] The early warning determination module is connected to the risk identification module and is used to generate corresponding chronic disease health supervision early warning results based on the health risk identification results and the traditional Chinese medicine constitution classification results.

[0012] Preferably, the chronic disease-related data further includes one or more of the following: behavioral data, medication data, past health record data, and user feedback data; The above data can be used to supplement physiological monitoring data with information reflecting the daily management of people with chronic diseases and individual differences, thus providing a more complete data foundation for chronic disease health supervision and early warning.

[0013] Preferably, the physiological monitoring data includes one or more of the following: blood pressure, blood glucose, heart rate, blood oxygen, body temperature, and weight. Physiological monitoring data can reflect the basic health status of people with chronic diseases from different perspectives, making it easier to select appropriate monitoring indicators to carry out health supervision and early warning according to different types of chronic diseases.

[0014] Preferably, the behavioral data includes one or more of the following: activity level, sleep patterns, dietary records, and exercise records; the medication data includes one or more of the following: medication time, medication frequency, missed doses, and medication adherence information. Behavioral and medication data can reflect the implementation of daily health management for individuals with chronic diseases, which is helpful in comprehensively considering the correlation between changes in health status and management implementation when identifying risks.

[0015] Preferably, the data processing module is further used to perform outlier removal, missing data completion, and unified dimensional processing on the chronic disease-related data; To reduce the impact of differences in data collection frequency, numerical format and data quality from different sources on the construction of subsequent regulatory features, and to improve the consistency and availability of basic regulatory data.

[0016] Preferably, the corresponding status data within the historical time period is the health status data of the chronic disease subject within a preset historical period; the stage change feature is used to characterize the degree of change in health status within the current time period relative to the health status within the preset historical period. By comparing the data of the current time period with the corresponding status data in the historical period, the early warning judgment can not only be based on the current status itself, but also reflect the phased changes in the health status of people with chronic diseases.

[0017] Preferably, the management execution data includes at least one or more of medication execution status, health behavior execution status, and follow-up execution status; the management features are used to characterize the execution status of chronic disease subjects in accordance with chronic disease management requirements; By incorporating management execution data into management characteristics and integrating them into the risk identification process, chronic disease health supervision and early warning can take into account both health status and management implementation, which is more in line with the actual needs of chronic disease continuous supervision scenarios.

[0018] Preferably, the early warning determination module is used to generate a graded early warning result based on the health risk identification result and the preset early warning rules, wherein the graded early warning result includes at least two levels among general reminder, key attention and high risk warning; By setting tiered early warning results, it is easier to adopt corresponding regulatory attention methods and early warning response strategies for different levels of risk.

[0019] Preferably, the system further includes a monitoring output module, which is connected to the early warning determination module and is used to output the graded early warning results to at least one of the user terminal, family member terminal, medical staff terminal, and management terminal, thereby meeting the needs of different monitored objects for receiving chronic disease health early warning information.

[0020] Preferably, the artificial intelligence model outputs a risk level or risk value as the health risk identification result; By outputting risk levels or risk values, the system can provide a clear basis for subsequent early warning judgments; at the same time, the system can be applied according to the health supervision needs of different chronic disease subjects.

[0021] The embodiments of the present invention have the following advantages due to the adoption of the above technical solutions: This invention comprehensively collects chronic disease-related data and TCM physical signs information from individuals with chronic diseases, and constructs current status characteristics, stage change characteristics, management characteristics, and constitution identification characteristics to achieve the fusion analysis of TCM constitution and modern health data; it improves the accuracy and personalization of chronic disease health supervision and early warning by using an artificial intelligence model for risk identification and combining it with constitution differences for comprehensive assessment.

[0022] The above overview is for illustrative purposes only and is not intended to be limiting in any way. In addition to the illustrative aspects, embodiments, and features described above, further aspects, embodiments, and features of the invention will become readily apparent from the accompanying drawings and the following detailed description. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a schematic diagram of the TCM constitution-based chronic disease health monitoring and early warning system based on artificial intelligence, as described in this invention. Detailed Implementation

[0025] In the following description, only certain exemplary embodiments are briefly described. As those skilled in the art will recognize, the described embodiments can be modified in various ways without departing from the spirit or scope of the invention. Therefore, the drawings and description are considered to be exemplary in nature and not restrictive.

[0026] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0027] like Figure 1 As shown in the figure, this embodiment of the invention provides an artificial intelligence-based TCM constitution-based chronic disease health monitoring and early warning system, used for continuous monitoring, risk identification, and early warning determination of the health status of individuals with chronic diseases. The system can be applied to family health management, primary care chronic disease follow-up, internet-based health management, community chronic disease monitoring, and other scenarios requiring continuous monitoring and early warning of individuals with chronic diseases.

[0028] In this embodiment, "chronic disease subjects" can be patients with hypertension, diabetes, coronary heart disease, chronic respiratory diseases, or other groups with long-term health monitoring needs.

[0029] The system in this embodiment includes a data acquisition module, a data processing module, a regulatory feature construction module, a risk identification module, and an early warning determination module; in some embodiments, it may also include a regulatory output module.

[0030] In one embodiment, the data acquisition module is used to collect chronic disease-related data from individuals with chronic diseases. The chronic disease-related data includes at least physiological monitoring data. The physiological monitoring data may include one or more of blood pressure, blood glucose, heart rate, blood oxygen, body temperature, and weight. Further, the chronic disease-related data may also include one or more of behavioral data, medication data, past health record data, and user feedback data.

[0031] The behavioral data may include one or more of the following: activity level, sleep patterns, dietary records, and exercise records; the medication data may include one or more of the following: medication time, medication frequency, missed doses, and medication adherence information; the past health record data may include one or more of the following: past disease information, physical examination information, follow-up visit information, and laboratory test information; the user feedback data may include information such as symptom changes, personal feelings, and descriptions of special events. The data acquisition module can obtain the relevant data through wearable devices, home monitoring devices, mobile terminals, health management terminals, medical information systems, or manual data entry.

[0032] In one embodiment, the data processing module is connected to the data acquisition module and is used to perform time alignment and normalization processing on chronic disease-related data to obtain regulatory basic data arranged in time series. Time alignment can be performed according to a preset time granularity, so that data belonging to the same time period corresponds to the same time window; normalization processing may include data unit unification, format conversion, numerical range unification, and label encoding. In a further embodiment, the data processing module can also perform outlier removal, missing data completion, and unit unification processing on the chronic disease-related data.

[0033] In one embodiment, the regulatory feature construction module is connected to the data processing module to construct a regulatory feature dataset based on regulatory foundation data. The regulatory feature dataset includes at least current state features, stage change features, and management features. Current state features reflect the health status of individuals with chronic diseases within the current time period; stage change features characterize the changes in health status within the current time period relative to historical time periods; and management features characterize the compliance status of individuals with chronic diseases with chronic disease management requirements.

[0034] Specifically, the regulatory feature construction module can use basic regulatory data within a preset time window as the current time period data, and extract current state features based on physiological monitoring data and related chronic disease data within this time period; then, it compares the current state features with the corresponding state data within historical time periods to form stage change features; simultaneously, it forms management features based on management execution data. The corresponding state data within the historical time period can be health state data of the same type or comparable to the current time period within a preset historical cycle; the management execution data includes at least one or more of medication execution, health behavior execution, and follow-up execution; and simultaneously, it extracts constitution identification-related features based on the TCM physical signs information for subsequent TCM constitution classification analysis.

[0035] In one embodiment, the risk identification module is connected to the regulatory feature construction module to input current status features, stage change features, and management features into an artificial intelligence model to obtain health risk identification results for chronic disease subjects. The artificial intelligence model can be a model suitable for health status identification, risk classification, or risk assessment, and its output can be a risk level or a risk value.

[0036] In one embodiment, the early warning determination module is connected to the risk identification module and is used to generate corresponding chronic disease health supervision early warning results based on the health risk identification results. Specifically, the early warning determination module can receive the risk category, risk level, abnormal indicator information and their changing trends output by the risk identification module, and comprehensively determine the relevant chronic disease health supervision risks in conjunction with the target user's current health status, thereby forming an early warning result corresponding to the target user. By setting up the early warning determination module, further rule-based processing of health risk identification results can be achieved, improving the timeliness of risk response and the accuracy of early warning output in the chronic disease health supervision process.

[0037] The early warning judgment module can determine the health risk identification results according to preset early warning rules to generate early warning results of corresponding levels. These preset early warning rules can be set based on at least one of the following: the degree of abnormality in health indicators, the duration of abnormality, the trend of historical health data changes, the type of chronic disease, concurrent risk factors, and the user's basic health record. Furthermore, the early warning judgment module can also perform tiered judgment of the target user's current chronic disease health risk based on the abnormality of a single indicator or the combined abnormality of multiple indicators, so that the generated early warning results better meet actual health supervision needs. Preferably, the early warning results are tiered early warning results, including at least two levels: general reminder, key attention, and high-risk warning. General reminders can be used to alert users to mild abnormalities or matters requiring daily health management attention; key attention can be used to alert users to persistent abnormalities, fluctuating abnormalities, or health risks with a further development trend; and high-risk warnings can be used to alert users to significant abnormalities, acute deterioration trends, or chronic disease risk situations requiring priority intervention. By setting up a tiered early warning mechanism, differentiated regulatory measures can be taken according to different risk levels, thereby improving the targeting and effectiveness of chronic disease health supervision.

[0038] In some embodiments, the system further includes a monitoring output module, which is connected to the early warning determination module and is used to output tiered early warning results to at least one of the user terminal, family terminal, medical staff terminal, and management terminal. The monitoring output module can output early warning results through methods such as message push, interface display, voice prompts, SMS notifications, or report generation, so that different receiving terminals can promptly obtain relevant health monitoring information. Furthermore, for different early warning levels, the monitoring output module can be configured with different output targets, output frequencies, and output content to meet various application needs such as personal health management, family collaborative monitoring, medical intervention, and institutional supervision.

[0039] Example 1: For individuals with hypertension and chronic disease, the data acquisition module collects blood pressure, heart rate, activity level, sleep patterns, medication time, and follow-up visit records. The data processing module aligns various data types by time, forming basic monitoring data arranged by day. The monitoring feature construction module extracts the daily average blood pressure, blood pressure fluctuations, and heart rate status as current status features, compares them with the blood pressure and heart rate data of the corresponding time periods in the previous seven days to form stage change features, and constructs management features based on the medication completion rate, exercise execution, and follow-up visit completion in the past seven days. The risk identification module inputs the above three types of features into the artificial intelligence model and outputs health risk identification results. If the identification results indicate that the individual with hypertension and chronic disease is in a state of continuously rising blood pressure and insufficient management execution, the early warning judgment module generates a key attention or high-risk early warning result, and extracts constitution identification features in combination with traditional Chinese medicine physical signs information.

[0040] Example 2: For individuals with chronic diabetes, the data collection module collects blood glucose, weight, dietary records, exercise records, medication adherence, and user feedback. The data processing module processes the collected data to form basic regulatory data. The regulatory feature construction module extracts blood glucose and weight status within the current time period as current status features, compares the current blood glucose-related status with corresponding status data within a preset historical period to form stage change features, and constructs management features based on dietary control, exercise, and medication adherence. The risk identification module outputs health risk identification results based on the above features. The early warning judgment module generates general reminders, key attention, or high-risk warnings based on the identification results and extracts constitution identification features from traditional Chinese medicine physical signs information for risk assessment.

[0041] Example 3: For individuals with chronic respiratory diseases, the data acquisition module collects information on blood oxygen saturation, heart rate, activity level, sleep patterns, medication adherence, and user feedback on symptoms such as chest tightness and shortness of breath. The data processing module generates basic regulatory data arranged in a time series. The regulatory feature construction module extracts blood oxygen saturation and activity status within the current time period as current status features, compares the current status with corresponding status data within a preset historical period to form stage change features, and generates management features based on medication adherence and follow-up visit adherence. The risk identification module integrates the above features and outputs health risk identification results. The early warning judgment module generates corresponding early warning results based on these results and combines traditional Chinese medicine physical signs information to extract constitution identification features for risk assessment.

[0042] In this embodiment, while identifying health risks, the artificial intelligence model can classify chronic disease subjects based on traditional Chinese medicine physical signs data, and modify or weight the risk identification results in combination with different physical types to improve the individual adaptability of the early warning results.

[0043] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various variations or substitutions within the technical scope disclosed in the present invention, and these should all be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A traditional Chinese medicine constitution chronic disease health monitoring and early warning system based on artificial intelligence, characterized in that, include: The module includes a health data and TCM vital signs collection module, a data processing module, a regulatory feature construction module, a risk identification module, and an early warning judgment module. The health data and TCM vital signs collection module is used to collect chronic disease-related data of chronic disease subjects. The chronic disease-related data includes at least physiological monitoring data and TCM vital signs information. The data processing module is connected to the health data and TCM vital signs collection module, and is used to perform time alignment and standardization processing on the chronic disease-related data to obtain basic regulatory data arranged in time series. The regulatory feature construction module is connected to the data processing module. It is used to extract current status features reflecting the health status of chronic disease subjects within the current time period based on the regulatory basic data, and compare the current status features with the corresponding status data in the historical time period to form stage change features representing the stage changes in health status. At the same time, it forms management features representing the implementation of chronic disease management based on the management execution data of chronic disease subjects, and extracts constitution identification-related features based on the TCM physical signs information to construct a regulatory feature dataset. The risk identification module is connected to the regulatory feature construction module and is used to input the current status features, stage change features, management features and constitution identification-related features into the artificial intelligence model to obtain the health risk identification results of chronic disease subjects and output the TCM constitution classification results. The early warning determination module is connected to the risk identification module and is used to generate corresponding chronic disease health supervision early warning results based on the health risk identification results and the traditional Chinese medicine constitution classification results.

2. The artificial intelligence-based traditional Chinese medicine constitution chronic disease health monitoring and early warning system according to claim 1, characterized in that, The chronic disease-related data also includes one or more of the following: behavioral data, medication data, past health record data, and user feedback data.

3. The artificial intelligence-based traditional Chinese medicine constitution chronic disease health monitoring and early warning system according to claim 1, characterized in that, The physiological monitoring data includes one or more of the following: blood pressure, blood glucose, heart rate, blood oxygen, body temperature, and weight.

4. The artificial intelligence-based traditional Chinese medicine constitution slow disease health monitoring and early warning system according to claim 2, characterized in that, The behavioral data includes one or more of the following: activity level, sleep patterns, dietary records, and exercise records; the medication data includes one or more of the following: medication time, medication frequency, missed doses, and medication adherence information.

5. The artificial intelligence-based traditional Chinese medicine constitution slow disease health monitoring and early warning system according to claim 1, characterized in that, The data processing module is also used to remove outliers, complete missing data, and standardize the units of measurement for the chronic disease-related data.

6. The artificial intelligence-based traditional Chinese medicine constitution slow disease health monitoring and early warning system according to claim 1, characterized in that, The corresponding status data within the historical time period is the health status data of the chronic disease subject within the same type of health status within the preset historical period; the stage change feature is used to characterize the degree of change in health status within the current time period relative to the health status within the preset historical period.

7. The artificial intelligence-based traditional Chinese medicine constitution slow disease health monitoring and early warning system according to claim 1, characterized in that, The management execution data includes at least one or more of the following: medication execution status, health behavior execution status, and follow-up execution status; the management features are used to characterize the execution status of chronic disease subjects in accordance with chronic disease management requirements. 8.The artificial intelligence-based traditional Chinese medicine constitution chronic disease health monitoring and early warning system according to claim 1, characterized in that, The early warning determination module is used to generate graded early warning results based on the health risk identification results and preset early warning rules. The graded early warning results include at least two levels: general reminder, key attention, and high-risk warning.

9. The artificial intelligence-based traditional Chinese medicine constitution slow disease health monitoring and early warning system according to claim 8, characterized in that, It also includes a monitoring output module, which is connected to the early warning determination module and is used to output the graded early warning results to at least one of the user terminal, family terminal, medical staff terminal, and management terminal.

10. The artificial intelligence-based traditional Chinese medicine constitution slow disease health monitoring and early warning system according to claim 1, characterized in that, The artificial intelligence model outputs a risk level or risk value as the health risk identification result.