Human body life health management method and system based on AI twinborn digital human

Abstract

The invention discloses a human body life health management method and system based on AI twinborn digital humans, and the method comprises the steps: collecting human body life health data in multiple dimensions, comprehensively and deeply collecting human body health information, breaking through the limitation of single traditional health management data, and providing a rich data basis for precise health analysis; through combination of an artificial intelligence model and a medical professional field knowledge base, deep mining and intelligent analysis can be performed on complex health data, multi-system / organ physiological age prediction, early risk prediction of diseases, accurate diagnosis and personalized health guidance are realized, and scientificity and effectiveness of health management are improved; based on the AI twinborn digital human generated by the user in a personalized manner, health management service is provided for the user in a visual and intimate manner, the sense of participation and acceptability of the user are enhanced, and the requirements of the user for personalized health service are met, so that the method has the advantages of high intelligent degree and good interaction effect.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a human life and health management method and system based on AI twin digital humans with high intelligence and good interactive effects. Background Art

[0002] With the rapid development of the Internet of Things (IoT), artificial intelligence (AI), and big data technologies, the healthcare sector is gradually transforming towards digitalization and intelligence. However, the existing technology system still faces the following key issues that need to be addressed: First, data silos and inefficient integration: Traditional healthcare data is scattered across various independent systems, lacking unified data collection standards and cross-platform integration capabilities. For example, a user's mental health assessment, metabolomics test results, and medical imaging data are often stored on different devices, resulting in poor data correlation, low comprehensive analysis efficiency, and difficulty in forming a comprehensive health picture.

[0003] Secondly, personalized service capabilities are insufficient: Existing health management systems rely on generalized models and lack the ability to deeply mine multidimensional user data. For example, there is no dynamic linkage mechanism between genetic testing, nutritional assessments, and exercise training plans. This makes it impossible to combine users' real-time physiological indicators (such as proteomics and cytology data) to provide precise intervention strategies, resulting in insufficiently targeted health management solutions.

[0004] Third, medical resources are unevenly distributed and the threshold for specialization is high: medical image analysis (such as lung CT and endoscopic images) is highly dependent on the experience of professional physicians, and primary medical institutions are short of resources, which can easily lead to diagnostic delays and the risk of misdiagnosis; ordinary users find it difficult to understand complex medical advice, and the existing interactive system lacks an intuitive and friendly communication interface, which reduces user compliance with health management.

[0005] Fourth, the user experience and interaction forms are single: traditional health management tools mainly use data reports or text suggestions, lack emotional and visual interactive design, and are particularly difficult to attract young user groups to actively participate in long-term health management.

[0006] In summary, although current technology can realize the collection and analysis of some health data, it has significant shortcomings in multi-source heterogeneous data fusion, personalized decision support, sinking of medical resources and user interaction experience. Therefore, it is necessary to propose an improvement to overcome the shortcomings of existing technology. Summary of the Invention

[0007] The purpose of this invention is to solve the problems in the existing technology and provide a human life and health management method and system based on AI twin digital humans. By comprehensively collecting human health data, conducting in-depth analysis with the help of advanced artificial intelligence algorithms, and providing users with accurate, intuitive, and easy-to-understand health management suggestions in the form of personalized AI twin digital humans, the efficiency and quality of health management are improved.

[0008] The technical solution of the present invention is: A human life and health management method based on AI twin digital humans comprises the following steps: S1. Collection of human life and health data: Collection of human life and health data of users in medical institutions, physical examination centers and daily life scenarios, and standardization of the collected human life and health data; S2. Data transmission and storage: Transmission of the standardized human life and health data to a cloud database through a network channel for storage, and formation of a medical professional knowledge base in combination with medical professional knowledge; S3. Data analysis and processing: Use an artificial intelligence model to conduct an in-depth analysis of the human life and health data in the cloud database to obtain user health results; S4. AI twin digital human presentation and interaction: Generate an AI twin digital human image based on photos or videos provided by the user; The AI ​​twin digital human interacts with the user based on user questions and the user health results obtained through analysis of the artificial intelligence model.

[0009] As a preferred technical solution, the standardized processing of the collected human life and health data in step S1 is specifically as follows: first, a data cleaning operation is performed to remove duplicate, erroneous, and incomplete data records; then, through a data association algorithm, data from different sources of the same user are integrated, and a complete user health data file is established with the user's unique identifier (such as ID number) as the index; during the data integration process, the data quality is evaluated, and for data that does not meet the quality standards or is questionable, a re-collection or manual review mechanism is automatically triggered.

[0010] As a preferred technical solution, in step S2, when storing human life and health data in a cloud database, the human life and health data are preprocessed, specifically: S2a, using data standardization methods to convert data of different dimensions and different value ranges into a unified standard form to facilitate subsequent model analysis; S2b, according to the characteristics and distribution of the data, using mean interpolation, regression interpolation, and multiple interpolation methods to interpolate missing data; S2c, detecting and processing abnormal data, identifying and marking outliers through statistical methods, correcting or deleting obviously erroneous outliers, and retaining and focusing on outliers that may reflect real physiological abnormalities.

[0011] As a preferred technical solution, the artificial intelligence model in step S3 includes a machine learning model, a deep learning model and a large language model.

[0012] As a further preferred technical solution, the use of an artificial intelligence model in step S3 to conduct in-depth analysis of human life and health data in the cloud database specifically includes the following steps: S31. Using a machine learning model and / or a deep learning model, and based on key features closely related to human life and health status, quantify the user's health status; S32. Combining the medical expertise in the medical professional knowledge base to perform pattern recognition and classification on human life and health data, and evaluate the user's health status; S33. Using a large language model intelligent agent based on the medical professional knowledge base to generate natural language according to the user's health status, health knowledge and suggestions, to form a user health interpretation result.

[0013] As a further preferred technical solution, quantifying the user's health status in step S31 specifically predicts the physiological age of the user's multiple systems or organs and predicts the user's risk of multiple diseases; evaluating the user's health status in step S32 specifically evaluates the user's health risks, suggests potential diseases, and analyzes health trends.

[0014] As a preferred technical solution, in step S4, the AI ​​twin digital human interacts with the user by understanding the user's questions through a large language model, and answers the user's questions in combination with the user's health results, medical knowledge and health advice. The AI ​​twin digital human presents the health results obtained by model analysis and the health management advice generated by the large language model intelligent agent to the user in various forms.

[0015] The present invention also provides a human life and health management system based on AI twin digital humans that implements the above-mentioned human life and health management method based on AI twin digital humans, including a human life and health data acquisition module, a cloud database, and an artificial intelligence module; the human life and health data acquisition module is used to collect human life and health data; the cloud database is used to store human life and health data and medical professional knowledge base; the artificial intelligence module is used to run artificial intelligence models and generate AI twin digital humans.

[0016] As a preferred technical solution, the human life and health data collection module includes one or more of a user basic information collection module, a questionnaire data collection module, a physical examination data collection module, a medical imaging data collection module, a lung detection data collection module, an endoscope detection data collection module, a genomics data collection module, a telomere length data collection module, a nutritional assessment data collection module, a mental health data collection module, a wearable device data collection module, a cytology detection data collection module, a proteomics detection data collection module, a metabolomics detection data collection module, a skin instrument data collection module, and a fundus camera data collection module.

[0017] As a preferred technical solution, the medical professional knowledge base includes one or more of a routine physical examination knowledge base, a gene decoding knowledge base, a diet nutrition knowledge base, a sports health knowledge base, a microbiology knowledge base, a mental health knowledge base, and a cytology knowledge base.

[0018] The human life and health management method and system based on AI twin digital humans of the present invention collect human life and health data in multiple dimensions, comprehensively and deeply collect human health information, break the limitation of the single traditional health management data, and provide a rich data foundation for accurate health analysis; through the combination of artificial intelligence models and medical professional knowledge bases, it can deeply mine and intelligently analyze complex health data, realize the physiological age prediction of multiple systems or organs, early risk prediction of diseases, accurate diagnosis and personalized health guidance, and improve the scientificity and effectiveness of health management; based on the AI ​​twin digital humans generated based on user personalization, it provides health management services to users in an intuitive and friendly manner, enhances the user's sense of participation and acceptance, and meets the user's demand for personalized health services. Therefore, the human life and health management method and system based on AI twin digital humans of the present invention have the advantages of high intelligence and good interactive effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is a flowchart of a specific implementation method of the human life and health management method based on AI twin digital humans of the present invention. DETAILED DESCRIPTION

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0021] The terms used in the embodiments of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention. The singular forms "a," "the," and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates otherwise. "A plurality" generally includes at least two, but does not exclude the inclusion of at least one.

[0022] It should be understood that the term "and / or" as used herein is merely a description of the relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " in this document generally indicates that the associated objects are in an "or" relationship.

[0023] As used herein, the words "if" and "if" may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to the determination" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)," depending on the context.

[0024] It should also be noted that the terms "include," "comprises," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a product or system comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such product or system. In the absence of further limitations, an element defined by the phrase "comprises a..." does not exclude the presence of other identical elements in the product or system comprising the element.

[0025] like Figure 1 The figure shows a specific implementation of a human life and health management method based on AI twin digital humans of the present invention. The human life and health management method based on AI twin digital humans of this embodiment includes the following steps: S1. Collection of human life and health data: Collect users' human life and health data in medical institutions, physical examination centers and daily life scenarios, and standardize the collected human life and health data. In order to collect human life and health data, users collect health data through various modules of the customer's IoT data collection system in medical institutions, physical examination centers, homes and other scenarios. For example, a comprehensive physical examination is conducted at a physical examination center, and medical imaging equipment, endoscope equipment, etc. are used to collect images and test data; users use wearable devices to record daily physiological data in real time at home; and fill in living habits and health-related information through online questionnaires. After preliminary quality control and format conversion, the collected data is standardized to ensure the accuracy and consistency of the data.

[0026] In this embodiment, in order to more comprehensively collect human life and health data, the following methods are used to collect human life and health data: Data Collection at Physical Examination Institutions: When users undergo various medical examinations at physical examination institutions, the corresponding equipment transmits the collected data in real time to the client's IoT data collection system. For example, after completing X-ray, CT, ultrasound, and MRI examinations, medical imaging equipment quickly and accurately transmits the image data to the system using the DICOM (Digital Imaging and Communications in Medicine) standard protocol. After completing tests, various testing instruments (such as blood testers and biochemical analyzers) encode and upload the test results in a unified data format (such as HL7, the Health Information Exchange Standard). Simultaneously, medical staff assist users in completing the relevant questionnaires in the questionnaire data collection module to ensure accurate entry of the information into the system.

[0027] Wearable device data collection: When users use wearable devices (smart bracelets, watches, blood pressure monitors, blood glucose meters, etc.) daily, the devices synchronize the collected real-time physiological data (such as heart rate, step count, blood pressure, and blood glucose) to the user's mobile terminal application via wireless communication technologies such as Bluetooth or WiFi. The mobile terminal application then uploads the data to the customer's IoT data collection system via a secure network connection. The system converts and normalizes the data from different brands and models of wearable devices to achieve effective data integration.

[0028] Other Test Data Collection: For specialized testing items such as genomics, proteomics, and metabolomics, after completing testing at a professional testing institution, the institution will send the test report in electronic form (such as PDF) to a designated email address or upload it to the system via an encrypted network interface. The system uses optical character recognition (OCR) and natural language processing technologies to extract and structure the data in the test report, enabling further analysis and utilization. Users can also upload data such as skin analyzer and fundus camera test reports through mobile terminal applications, and the system will also perform corresponding data processing and integration operations.

[0029] S2. Data Transmission and Storage: Normalized human health data is transmitted via network channels to a cloud database for storage and integrated with medical expertise to form a medical expertise database. During the data transmission and storage phase, normalized data is transmitted via secure network channels to a data processing system based on the AI ​​digital human. The data processing system stores the data in a cloud database, utilizing distributed storage and backup strategies to ensure data security and reliability while facilitating subsequent data queries and access.

[0030] S3. Data analysis and processing: An artificial intelligence model is used to conduct an in-depth analysis of the human life and health data in the cloud database to obtain the user's health results. In this embodiment, the machine learning / deep learning model in the data processing and the large model agent perform in-depth analysis of the stored health data. First, the model quantifies the user's health status based on key features closely related to the human life and health status. Then, combined with the professional knowledge in the knowledge base, the data is pattern recognized and classified to assess the user's health risks, diagnose potential diseases, analyze health trends, etc. Based on the model analysis results, the large language model agent extracts relevant health knowledge and suggestions from the knowledge base, generates natural language, and converts professional medical knowledge into easy-to-understand expressions. The constructed machine learning or deep learning model is used to deeply analyze the input data. Based on the model analysis results, the large language model agent combines multi-domain knowledge bases for knowledge fusion and reasoning. For example: When the machine learning / deep learning model outputs that the user is at risk of prediabetes, the large language model agent extracts dietary recommendations suitable for people with prediabetes (such as controlling carbohydrate intake, increasing dietary fiber intake, etc.) from the dietary nutrition knowledge base, extracts suitable exercise plans (such as the frequency, intensity and duration of aerobic exercise, etc.) from the sports health knowledge base, and extracts psychological adjustment methods for coping with disease risks (such as techniques for relieving anxiety, etc.) from the mental health knowledge base. Through knowledge reasoning and semantic understanding, it integrates this knowledge into a complete, coherent, and easy-to-understand set of health management recommendations.

[0031] S4. AI Twin Digital Human Presentation and Interaction: An AI twin digital human is generated based on user-provided photos or videos. The AI ​​twin interacts with the user based on their questions and the user's health results analyzed by the AI ​​model. Based on the user's photos or videos, the visual large model quickly generates an AI twin digital human. The AI ​​twin vividly presents the health results (such as disease risk assessment level, abnormal physical examination indicators, etc.) and related recommendations (such as dietary adjustment plans, exercise plans, and mental health adjustment methods) derived from the model analysis to the user. Users can interact with the digital human and ask further questions or requests. The digital human, through the large language model agent, understands the user's needs and provides timely and accurate responses, achieving a personalized and interactive health management service experience.

[0032] In this embodiment, a human life and health management method based on AI twin digital humans is implemented. In step S1, the collected human life and health data are standardized as follows: first, data cleaning operations are performed to remove duplicate, erroneous, and incomplete data records; then, data from different sources of the same user are integrated through a data association algorithm, and a complete user health data file is established with the user's unique identifier (such as an ID number) as an index; during the data integration process, the data quality is assessed, and for data that does not meet the quality standards or is questionable, a re-collection or manual review mechanism is automatically triggered.

[0033] In this embodiment, a method for managing human life and health based on AI twin digital humans is provided. In step S2, when storing human life and health data in a cloud database, the human life and health data is pre-processed, specifically: S2a, using data standardization methods to convert data of different dimensions and value ranges into a unified standard form to facilitate subsequent model analysis; S2b. Based on the characteristics and distribution of the data, the missing data are interpolated using mean interpolation, regression interpolation, and multiple interpolation methods; S2c. Detect and process abnormal data. Outliers are identified and labeled using statistical methods. Clearly erroneous outliers are corrected or deleted, while those that may reflect true physiological abnormalities are retained and prioritized. In specific applications, after collecting and integrating user health data, data preprocessing is first performed. Data standardization methods are used to convert data of varying dimensions and value ranges into a unified, standardized format to facilitate subsequent model analysis. Missing data are also interpolated. Depending on the data's characteristics and distribution, methods such as mean interpolation, regression interpolation, and multiple interpolation can be used. For example, for occasionally missing heart rate data from some wearable devices, linear regression can be used to interpolate heart rate data from adjacent time points. For a small number of missing laboratory test parameters in physical examination data, the mean value of the parameter within the same population can be used for interpolation. Furthermore, abnormal data are detected and processed. Outliers are identified and labeled using statistical methods (such as the 3σ principle and boxplots). Clearly erroneous outliers are corrected or deleted, while those that may reflect true physiological abnormalities are retained and prioritized. It should be noted that in specific applications, steps S2a, S2b and S2c can be flexibly selected according to actual needs. There is no corresponding order relationship among steps S2a, S2b and S2c, and not all of them need to be applied.

[0034] In this embodiment, a method for managing human life and health based on an AI twin digital human is described. The artificial intelligence model described in step S3 includes a machine learning model, a deep learning model, and a large language model. The machine learning / deep learning model utilizes a large amount of accumulated data for training. Through techniques such as feature extraction and pattern recognition, it conducts in-depth analysis of collected health data, enabling functions such as physiological age prediction, disease risk prediction, disease diagnosis assistance, and health trend analysis. The large language model possesses powerful natural language understanding and generation capabilities. It can understand user input for health-related questions, extract relevant knowledge from a knowledge base, and, through natural language generation technology, provide professional health advice and recommendations in plain, logical language. For example, when a user asks, "I've been feeling tired lately. What might be the cause?" the large language model agent can comprehensively analyze the user's physical examination data, lifestyle habits, and other information, answering questions from multiple perspectives, such as diet and nutrition, lack of exercise, psychological stress, and potential medical conditions, and providing targeted improvement suggestions.

[0035] In step S3, the artificial intelligence model is used to perform in-depth analysis of the human life and health data in the cloud database, which specifically includes the following steps: S31. Use machine learning models and / or deep learning models to quantify the user's health status based on key characteristics closely related to human life and health status; S32. Combining medical expertise in the medical expertise database, perform pattern recognition and classification on human life and health data to assess the user's health status; S33. Use a large language model agent based on a medical professional knowledge base to generate natural language based on the user's health status, health knowledge and suggestions to form a user health interpretation result.

[0036] Quantifying the user's health status in step S31 specifically involves predicting the physiological age of multiple systems or organs of the user and predicting the user's risk of multiple diseases; evaluating the user's health status in step S32 specifically involves evaluating the user's health risks, indicating potential diseases, and analyzing health trends.

[0037] In this embodiment, step S3 uses a pre-built machine learning or deep learning model to deeply analyze the input data. Based on the model analysis results, the large language model agent combines multi-domain knowledge bases to perform knowledge fusion and reasoning. For example, if the machine learning / deep learning model indicates that a user is at risk for prediabetes, the large language model agent extracts dietary recommendations suitable for people with prediabetes (such as controlling carbohydrate intake and increasing dietary fiber intake) from the dietary nutrition knowledge base, appropriate exercise plans (such as the frequency, intensity, and duration of aerobic exercise) from the sports health knowledge base, and psychological adjustment methods for coping with disease risks (such as techniques for alleviating anxiety) from the mental health knowledge base. Through knowledge reasoning and semantic understanding, the large language model agent integrates this knowledge into a complete, coherent, and easily understood set of health management recommendations.

[0038] In this embodiment, a method for managing human life and health based on an AI twin digital human is described. In step S4, the AI ​​twin digital human interacts with the user by understanding the user's question using a large language model and providing answers based on the user's health results, medical knowledge, and health recommendations. The AI ​​twin digital human presents the health results analyzed by the model and the health management recommendations generated by the large language model agent to the user in various formats. The AI ​​twin digital human appears in animated form on a mobile terminal application interface or WeChat mini-program, providing a detailed explanation of the user's health status through voice broadcasts, such as "Your current blood pressure is critical. You need to pay attention to your salt intake in your daily diet." For complex medical concepts and health recommendations, the digital human uses animations to assist in explanation. For example, when explaining exercise recommendations, animations demonstrate standard exercise movements and rhythms. Users can also view text-based health reports within the application, which include detailed physical examination indicator analysis, disease risk assessment results, and personalized health management plans. Users can also interact with the digital human in real time by inputting questions through voice or text, such as "How long will it take for me to see results if I follow this exercise plan?" The digital human will understand the user's question through the large language model intelligent agent and quickly give accurate and easy-to-understand answers.

[0039] This embodiment also provides a human life and health management system based on AI twin digital humans, including a human life and health data acquisition module, a cloud database, and an artificial intelligence module; the human life and health data acquisition module is used to collect human life and health data; the cloud database is used to store human life and health data and medical professional knowledge base; the artificial intelligence module is used to run artificial intelligence models and generate AI twin digital humans.

[0040] The human life and health data collection module is the hardware equipment used to deploy customer IoT data collection systems in various medical institutions (hospitals, physical examination centers), community health service stations, and other locations. The module includes medical imaging equipment, endoscopes, testing instruments (such as blood testers and trace element detectors), and wearable device data receiving terminals. Furthermore, corresponding mobile applications or mini-programs are developed to facilitate users to complete questionnaires, upload personal photos or videos, and access health data and services via mobile devices such as mobile phones.

[0041] In this embodiment, a human life and health management system based on AI twin digital humans is provided. The human life and health data collection module includes one or more of a user basic information collection module, a questionnaire data collection module, a physical examination data collection module, a medical imaging data collection module, a lung detection data collection module, an endoscope detection data collection module, a genomics data collection module, a telomere length data collection module, a nutritional assessment data collection module, a mental health data collection module, a wearable device data collection module, a cytology detection data collection module, a proteomics detection data collection module, a metabolomics detection data collection module, a skin instrument data collection module, and a fundus camera data collection module. Specifically: The basic customer information collection module is primarily responsible for collecting basic user identity information, including but not limited to name, age, gender, height, weight, contact information, and occupation. This basic information is an important basis for subsequent health analysis and management, and can provide foundational data for analyzing the health characteristics of different user groups. For example, age and gender are important factors in disease risk assessment, and different occupational groups may have specific health risk factors.

[0042] The questionnaire data collection module collects information about the user's lifestyle and health through questionnaires. This includes information on dietary habits (such as preference for high-salt, high-sugar, and high-fat foods, number of meals per day, etc.), exercise habits (weekly frequency, type of exercise, and duration of exercise), sleep patterns (bedtime, sleep duration, and self-assessed sleep quality), smoking and drinking history (years of smoking, daily cigarette count, frequency and amount of drinking), family medical history (whether there are hereditary diseases such as diabetes, hypertension, and cancer in the family), and past medical history. This information allows for a preliminary assessment of the user's health behaviors and potential health risks.

[0043] Physical Examination Data Collection Module: This module collects data from routine physical examinations. Blood pressure data reflects the stress on the cardiovascular system and is crucial for screening for conditions like hypertension. Body composition analysis measures body fat, muscle mass, and water content, used to assess nutritional status and metabolic levels. Routine blood tests, including red blood cell (RBC), white blood cell (WBC), and platelet counts, can reflect the body's basic hematopoietic function, immune status, and the presence of infection, anemia, and other conditions.

[0044] Medical imaging data acquisition modules include: 1. X-ray detection: X-rays use their penetrating properties to image bones, lungs, and other parts of the body. This can be used to detect fractures, lung inflammation, lung tumors, and other diseases.

[0045] 2. CT scan: By performing cross-sectional scans on the human body, it provides more detailed images of the internal structure and is of great value in the diagnosis of brain diseases (such as brain tumors and cerebral hemorrhage), lung diseases (such as lung cancer and lung nodules), and abdominal diseases (such as liver tumors and kidney stones).

[0046] 3. Ultrasound testing: Using ultrasound principles, it examines the heart, blood vessels, abdominal organs (such as the liver, gallbladder, pancreas, and kidneys), and thyroid gland. It can detect abnormalities in heart structure and function, vascular plaques, and changes in organ morphology and function.

[0047] 4. MRI testing: Based on the principle of nuclear magnetic resonance, it can provide high-resolution soft tissue images and plays an important role in the diagnosis of neurological diseases (such as cerebral infarction, brain tumors, spinal cord lesions), musculoskeletal system diseases (such as muscle strains, joint lesions), and breast diseases.

[0048] Lung detection data acquisition module, including: 1. Exhaled breath composition testing: Analyzing the chemical composition of exhaled breath, such as nitric oxide and carbon monoxide, helps assess the risk of diseases such as airway inflammation and lung infection. For example, elevated nitric oxide levels in exhaled breath may indicate airway inflammation.

[0049] 2. Pulmonary function test: By measuring indicators such as forced vital capacity, forced expiratory volume in one second, and maximum ventilation, the ventilation function of the lungs is assessed. It can be used to diagnose lung diseases such as chronic obstructive pulmonary disease, asthma, and emphysema.

[0050] Endoscopic detection data acquisition module, including: 1. Esophageal / gastroscopy: Inserting an endoscope into the esophagus and stomach allows direct observation of the morphology, color, presence of ulcers, tumors, and other lesions of the esophageal and gastric mucosa. This is crucial for the early diagnosis of diseases such as esophageal cancer, gastric cancer, and gastric ulcers.

[0051] 2. Colonoscopy: Used to observe the internal conditions of the intestines, it can detect intestinal polyps, enteritis, intestinal cancer and other lesions. It is an important means of screening and diagnosing intestinal diseases.

[0052] Genomics data acquisition module, including: 1. Genetic risk testing: By testing specific gene loci, the user's risk of developing certain genetic diseases (such as hereditary breast cancer, ovarian cancer, and hereditary cardiovascular disease) is analyzed. For example, testing for BRCA1 and BRCA2 gene mutations can assess the genetic risk of breast cancer and ovarian cancer.

[0053] 2. ctDNA cancer early screening: By detecting circulating tumor DNA (ctDNA) in the blood, multiple cancer-related genes are tested to assess an individual's risk of developing a certain cancer, which helps with early screening and prevention.

[0054] 3. Intestinal microbial testing: Analyze the composition and diversity of intestinal microbial communities. Intestinal microbes are closely related to human health, and their imbalance may be related to various diseases such as obesity, diabetes, and inflammatory bowel disease.

[0055] Telomere Length Data Collection Module: Telomeres are protective structures at the ends of chromosomes that gradually shorten as cells divide. Telomere length measurement can, to a certain extent, reflect the degree of cellular aging and is associated with the risk assessment of aging-related diseases (such as cardiovascular disease and neurodegenerative diseases).

[0056] Nutritional assessment data collection module, including: 1. Trace element testing: Detects the content of trace elements such as iron, zinc, copper, magnesium, and calcium in the human body. These elements play an important role in human physiological functions. Deficiency or excess of these elements may affect health. For example, iron deficiency can lead to anemia, and zinc deficiency can affect growth and development and immune function.

[0057] 2. Vitamin testing: Measures the levels of vitamin A, B vitamins, vitamin C, vitamin D, vitamin E, etc. Vitamin deficiency or excess can also cause various health problems. For example, vitamin D deficiency can lead to osteoporosis, and vitamin C deficiency can cause scurvy.

[0058] 3. Mineral testing: This test assesses the levels of minerals such as sodium, potassium, and chloride, which are essential for maintaining the body's water and salt balance, neuromuscular excitability, and other physiological functions. For example, sodium-potassium imbalance may affect blood pressure and heart function.

[0059] Mental health data collection module, including: 1. Psychological scale: Use professional psychological assessment scale to evaluate the user's psychological state from multiple dimensions and assess the customer's psychological state.

[0060] 2. EEG monitor: By detecting the brain's electrical activity and analyzing changes in different brainwave frequency bands, the user's brain activity status can be understood, and the user's mental state, sleep quality, and concentration level can be judged.

[0061] 3. Facial Expression Recognition: Using computer vision technology, it analyzes subtle changes in the user's facial expressions and identifies emotional states as an auxiliary means of mental health assessment.

[0062] Wearable device data acquisition module, including: 1. Smart bracelets / watches: Real-time monitoring of heart rate, steps, distance traveled, calorie consumption, sleep quality (including the duration and transitions between light sleep, deep sleep, and REM sleep stages), and other data, providing convenient dynamic physiological information for daily health management.

[0063] 2. Blood pressure monitoring: Wearable blood pressure monitoring devices can measure blood pressure regularly or in real time, allowing users to monitor blood pressure changes at home. This is of great significance for blood pressure management in patients with hypertension.

[0064] 3. Blood glucose monitoring: Some wearable blood glucose monitoring devices monitor blood glucose levels in real time through non-invasive or minimally invasive methods, helping diabetic patients better control their blood glucose and promptly detect abnormal blood glucose fluctuations.

[0065] 4. ECG monitoring: Wearable ECG devices can record electrocardiograms and detect the heart's electrical activity, which is of great value for screening and monitoring arrhythmias (such as premature beats and atrial fibrillation).

[0066] 5. Sleep monitoring: In addition to the sleep monitoring function of smart bracelets / watches, there are also dedicated sleep monitoring devices that can more accurately monitor sleep breathing conditions (such as whether there is sleep apnea), body movements, etc., and comprehensively assess sleep quality.

[0067] Cytology detection data acquisition module, including: 1. Immune cell testing: This test measures the number, activity, and functional status of immune cells (such as T cells, B cells, and NK cells) to assess the function of the human immune system. Abnormal immune function is closely related to the occurrence and development of various diseases (such as infectious diseases, autoimmune diseases, and tumors).

[0068] 2. Circulating tumor cell detection: By detecting the number and characteristics of circulating tumor cells in the blood, it helps in early diagnosis, disease monitoring and prognosis assessment of tumors. Especially for patients with solid tumors, the presence of tumor cells can be detected before the tumor forms obvious symptoms.

[0069] Proteomics Data Acquisition Module: This module detects the expression of the human proteome and analyzes protein types, content, and modification status. Proteomics data can be used to discover disease biomarkers. For example, abnormal expression of certain tumor marker proteins can indicate tumor development. Proteomics data can also be used to evaluate drug efficacy and study disease pathogenesis.

[0070] Metabolomics Data Acquisition Module: Analyzes the types and content changes of human metabolites. Metabolomics can reflect the body's metabolic state and is closely related to the occurrence and development of diseases and drug metabolism. For example, by detecting metabolites in blood or urine, it can assist in the diagnosis of metabolic diseases such as diabetes and obesity, and can also be used to evaluate the effectiveness of disease treatments and drug safety.

[0071] Skin meter data acquisition module: The skin meter detects indicators such as skin moisture content and pigmentation to evaluate the health and aging of the skin, providing a scientific basis for skin care and beauty, and also assisting in the auxiliary diagnosis of certain skin diseases (such as seborrheic dermatitis, chloasma, etc.).

[0072] Fundus camera data acquisition module: This module uses a fundus camera to capture fundus images and observe the morphology and vascular conditions of the retina and other areas. Fundus lesions are closely related to various systemic diseases (such as diabetes, hypertension, and glaucoma). Fundus image analysis can help detect ocular complications of these diseases early and also help assess disease progression.

[0073] In this embodiment, a human life and health management system based on AI twin digital humans is provided, wherein the medical professional knowledge base includes one or more of a routine physical examination knowledge base, a gene decoding knowledge base, a diet and nutrition knowledge base, a sports and health knowledge base, a microbiology knowledge base, a mental health knowledge base, and a cytology knowledge base. Specifically: Routine Physical Examination Knowledge Base: This integrates knowledge on various routine physical examination items, including testing methods, normal reference ranges, interpretation of abnormal results, and their association with diseases. For example, if a white blood cell count in a blood test is abnormally elevated, the knowledge base will detail possible causes, such as infection (bacterial, viral, etc.), inflammatory response, or blood system disease, and provide further examination recommendations.

[0074] Gene decoding knowledge base: This contains information on gene structure, function, genetic variation, and the relationship between genes and diseases. Taking the BRCA1 gene mutation as an example, the knowledge base explains the relationship between this mutation and the risk of breast and ovarian cancer, as well as prevention and intervention measures for individuals with this mutation.

[0075] Dietary nutrition knowledge base: covers the nutritional components of various foods, nutritional needs of different groups of people, dietary matching principles, and dietary intervention plans for nutrition-related diseases.

[0076] Sports Health Knowledge Base: Covers knowledge of exercise physiology, such as the effects of different types of exercise (aerobic exercise, anaerobic exercise) on various body systems (cardiovascular system, musculoskeletal system, etc.); principles for formulating exercise prescriptions, including the determination of exercise intensity, frequency, time and method; and prevention and treatment methods for sports injuries.

[0077] Microbial knowledge base: includes the composition, function, relationship with health and disease of human microbial communities (such as intestinal microorganisms, oral microorganisms, etc.), and methods to regulate the balance of microbial communities (such as the role of probiotics, the impact of diet on microorganisms, etc.).

[0078] Mental Health Knowledge Base: This database contains information on the symptoms, diagnostic criteria, pathogenesis, treatments, and methods for maintaining and promoting mental health (such as psychological adjustment techniques and stress management strategies) of common mental illnesses. It also covers core topics such as personality, well-being, and social support. By integrating this knowledge, it provides a theoretical basis for mental health assessment and intervention, helping users improve their mental state and achieve comprehensive health management.

[0079] The Cytology Knowledge Base covers basic cytological knowledge, including cell structure, function, cell cycle, cell differentiation, and apoptosis, as well as the mechanisms of cell changes during disease development (such as the characteristics of tumor cells and the role of immune cells in tumor immunity) and treatment. It also covers stem cell classification (e.g., embryonic stem cells, adult stem cells), characteristics (self-renewal, multipotential differentiation), applications in disease treatment (e.g., for Parkinson's disease and leukemia), and related clinical research progress. Natural killer (NK) cell intervention covers the immune functions of NK cells (anti-tumor and anti-viral), activation and expansion techniques, and application cases and potential risks in immunotherapy, enriching the knowledge base for cytological health management.

[0080] This embodiment presents a human life and health management method and system based on an AI twin digital human. This method and system comprehensively collects human health data, conducts in-depth analysis using advanced artificial intelligence algorithms, and provides users with accurate, intuitive, and easy-to-understand health management recommendations in the form of a personalized AI twin digital human, improving the efficiency and quality of health management. This method and system, based on an AI twin digital human, uses IoT devices to uniformly collect multi-dimensional health data, breaking down the barriers of traditional medical data scattered across different devices and systems and enabling centralized management and standardized processing of cross-domain data. Combining machine learning, deep learning, and a large language model agent, it integrates multi-domain knowledge bases to provide comprehensive and accurate health assessments, risk predictions, and customized recommendations (such as disease prevention, nutritional intervention, exercise intervention, and psychological counseling). Through natural language interaction between the AI ​​twin doctor digital human and the AI ​​twin doctor, the system lowers the barrier to professional medical advice for users. The user-customized AI twin digital human makes health management more accessible and engaging, particularly for students and young users, improving user engagement and long-term health management compliance.

[0081] Compared to existing technologies, the human life and health management method and system based on an AI twin digital human in this embodiment has the following advantages: 1. Comprehensive data: The client's IoT data collection system includes multi-dimensional health data collection modules, enabling comprehensive and in-depth collection of human health information, breaking the limitations of traditional health management data that is limited to a single entity, and providing a rich data foundation for accurate health analysis. 2. Intelligent analysis and prediction: The machine learning / deep learning models and large language model agents in the AI ​​digital human-based data processing system, combined with a multi-domain knowledge base, can conduct in-depth mining and intelligent analysis of complex health data, enabling early disease risk prediction, accurate diagnosis, and personalized health guidance, thereby improving the scientific nature and effectiveness of health management. 3. Personalized service experience: The AI ​​twin digital human generated based on the user's personal characteristics provides users with health management services in an intuitive and friendly manner, enhancing user engagement and acceptance, and meeting users' demand for personalized health services. 4. Convenience and interactivity: Through IoT technology and cloud-based data processing, users can collect and query health data anytime, anywhere, interact with the AI ​​twin digital human in real time, and obtain professional health advice, making health management more convenient and efficient.

[0082] The above description is only a preferred embodiment of the present invention and is not intended to limit the scope of the present invention. That is, any equivalent changes and modifications made according to the content of the patent application of the present invention should fall within the technical scope of the present invention.

Claims

1. A human life and health management method based on AI twin digital humans, characterized by: The following steps are involved: S1. Collecting human life and health data: Collecting users' human life and health data in medical institutions, physical examination centers and daily life scenarios, and standardizing the collected human life and health data; S2. Data transmission and storage: The standardized human life and health data is transmitted to the cloud database through the network channel for storage, and combined with medical professional knowledge to form a medical professional knowledge database; S3. Data analysis and processing: Use artificial intelligence models to conduct in-depth analysis of human life and health data in the cloud database to obtain user health results; S4. AI twin digital human presentation and interaction: Generate an AI twin digital human image based on photos or videos provided by the user; the AI ​​twin digital human interacts with the user based on user questions and user health results obtained through analysis of the artificial intelligence model.

2. The human life and health management method based on AI twin digital human according to claim 1 is characterized by: In step S1, the collected human life and health data are standardized as follows: first, data cleaning operations are performed to remove duplicate, erroneous, and incomplete data records; then, through the data association algorithm, data from different sources of the same user are integrated, and the user's unique identifier (such as ID number) is used as the index to establish a complete user health data file; during the data integration process, the data quality is evaluated, and for data that does not meet the quality standards or is questionable, a re-collection or manual review mechanism is automatically triggered.

3. The human life and health management method based on AI twin digital human according to claim 1 is characterized by: When storing human life and health data in the cloud database in step S2, the human life and health data is pre-processed, specifically: S2a, using data standardization methods to convert data of different dimensions and value ranges into a unified standard form to facilitate subsequent model analysis; S2b. Based on the characteristics and distribution of the data, the missing data are interpolated using mean interpolation, regression interpolation, and multiple interpolation methods; S2c. Detect and process abnormal data, identify and mark outliers through statistical methods, correct or delete obviously erroneous outliers, and retain and focus on outliers that may reflect real physiological abnormalities.

4. The human life and health management method based on AI twin digital human according to claim 1 is characterized by: The artificial intelligence model described in step S3 includes a machine learning model, a deep learning model and a large language model.

5. The human life and health management method based on AI twin digital human according to claim 4 is characterized by: In step S3, the artificial intelligence model is used to perform in-depth analysis of the human life and health data in the cloud database, which specifically includes the following steps: S31. Use machine learning models and / or deep learning models to quantify the user's health status based on key characteristics closely related to human life and health status; S32. Combining medical expertise in the medical expertise database, perform pattern recognition and classification on human life and health data to assess the user's health status; S33. Use a large language model agent based on a medical professional knowledge base to generate natural language based on the user's health status, health knowledge and suggestions to form a user health interpretation result.

6. The human life and health management method based on AI twin digital human according to claim 5 is characterized by: Quantifying the user's health status in step S31 specifically involves predicting the physiological age of the user's multiple systems or organs and predicting the user's risk of multiple diseases; evaluating the user's health status in step S32 specifically involves evaluating the user's health risks, suggesting potential diseases, and analyzing health trends.

7. The human life and health management method based on AI twin digital human according to claim 1 is characterized by: In step S4, the AI ​​twin digital human interacts with the user by understanding the user's questions through the large language model, and answers the user's questions in combination with the user's health results, medical knowledge and health recommendations. The AI ​​twin digital human presents the health results obtained by model analysis and the health management recommendations generated by the large language model intelligent agent to the user in various forms.

8. A human life and health management system based on an AI twin digital human that implements the human life and health management method based on an AI twin digital human as described in claims 1-7, characterized in that: It includes a human life and health data acquisition module, a cloud database, and an artificial intelligence module; the human life and health data acquisition module is used to collect human life and health data; the cloud database is used to store human life and health data and medical professional knowledge base; the artificial intelligence module is used to run artificial intelligence models and generate AI twin digital people.

9. The human life and health management system based on AI twin digital human according to claim 8 is characterized by: The human life and health data collection module includes one or more of a user basic information collection module, a questionnaire data collection module, a physical examination data collection module, a medical imaging data collection module, a lung detection data collection module, an endoscope detection data collection module, a genomics data collection module, a telomere length data collection module, a nutritional assessment data collection module, a mental health data collection module, a wearable device data collection module, a cytology detection data collection module, a proteomics detection data collection module, a metabolomics detection data collection module, a skin instrument data collection module, and a fundus camera data collection module.

10. The human life and health management system based on AI twin digital human according to claim 8 is characterized by: The medical professional knowledge base includes one or more of a routine physical examination knowledge base, a gene decoding knowledge base, a diet nutrition knowledge base, a sports health knowledge base, a microbiology knowledge base, a mental health knowledge base, and a cytology knowledge base.

Images