Physical examination system based on medical data analysis

Abstract

The invention discloses a physical examination system based on medical data analysis, and belongs to the technical field of medical data processing and health management. The system comprises a data acquisition module which acquires user physical examination data including a user ID, physical examination time, a physical examination result and a physical examination basic index through an HL7FHIR protocol; the data integration module constructs a personal health basic archive set based on user physical examination data, and constructs a group disease factor model library containing n diseases and one health index mapping model through dynamic density segmentation processing; the intelligent analysis module performs similarity analysis on the user physical examination basic indexes and a group disease factor model library, generates a personal disease risk probability sequence and judges a risk level to obtain a user physical examination disease set; and the medical module matches a physical examination screening item scheme based on the user physical examination disease set and pushes the scheme to the physical examination user. According to the system, accurate risk assessment and targeted screening of diseases are realized, and the physical examination efficiency and the health management level are improved.

Description

Technical Field

[0001] This invention belongs to the field of medical data processing and health management technology, specifically relating to a health check-up system based on medical data analysis. Background Technology

[0002] With increased health awareness and advancements in medical technology, regular physical examinations to detect potential health risks have become an important means of disease prevention and health maintenance. However, current health check-up systems still have many shortcomings in practical applications, making it difficult to meet the needs of precise and personalized health management.

[0003] First, there are barriers to data collection and integration. Existing health checkup systems mostly rely on manual data entry or data exchange between a single hospital information system (HIS) and a laboratory information system (LIS), lacking a unified data interaction standard. This makes it difficult to integrate health checkup data from different institutions and at different times, resulting in insufficient continuity and completeness of personal health data and failing to provide effective support for long-term health assessment. Second, disease risk assessment lacks scientific rigor and accuracy. Traditional health checkup risk assessments often rely on doctors' experience or fixed threshold comparisons, failing to fully utilize group medical data to build quantitative models. Due to individual differences and the complexity of disease correlations, relying solely on a single indicator or experience can easily lead to misjudgments, making it difficult to accurately identify potential disease risks and thus affecting the targeted nature of intervention measures. Third, health checkup screening items lack personalization. Existing health checkup plans are mostly standardized packages that are not dynamically adjusted based on individual health indicators and disease risk characteristics. This leads to unnecessary examinations, increasing the cost and burden of health checkups, and may also miss targeted screenings for high-risk diseases, reducing the efficiency of health checkups and the effectiveness of health management.

[0004] Therefore, how to build a health check-up system based on standardized data collection, population model analysis, personalized risk assessment, and screening programs to improve the accuracy and efficiency of health check-ups has become a pressing technical problem in the field of medical and health management. Summary of the Invention

[0005] The purpose of this invention is to provide a health check-up system based on medical data analysis to solve the problems mentioned in the background art.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: A health check-up system based on medical data analysis includes a data acquisition module, a data integration module, an intelligent analysis module, and a medical module; The data acquisition module is used to acquire all users' physical examination data; The data integration module is used to construct a set of basic personal health records based on all users' physical examination data and store them in the database, and to construct a group disease factor model library based on the set of basic personal health records. The intelligent analysis module is used to perform similarity analysis between the user's basic physical examination indicators and the health indicator mapping model and various disease indicator mapping models in the group disease factor model library to obtain an individual disease risk probability sequence, and to judge the risk level of the individual disease risk probability sequence to obtain the user's physical examination disease set. The medical module is used by doctors to push the selected user's basic physical examination indicators to the intelligent analysis module. At the same time, it matches the physical examination screening program plan based on the user's physical examination disease set, and pushes the individual disease risk probability sequence and physical examination screening program plan to the physical examination user to which the user's basic physical examination indicators belong via SMS.

[0007] Preferably, the user's physical examination data is obtained by connecting to the hospital's Hospital Information System (HIS) and Laboratory Information System (LIS) via the HL7FHIR protocol. The HL7FHIR protocol, as a common data interaction standard in the medical and health field, enables standardized transmission and integration of data between different systems. The user's physical examination data includes user ID, examination time, examination results, and basic physical examination indicators. The user ID is generated by irreversibly encrypting the user's electronic medical insurance voucher code, electronic health card number, or ID card number using the SHA-256 encryption algorithm. The examination results are categorized as either healthy or n diseases, where n is a positive integer not less than 1, and each disease corresponds to a specific disease type. The basic physical examination indicators include Body Mass Index (BMI), fasting blood glucose, systolic blood pressure, diastolic blood pressure, total cholesterol, triglycerides, hemoglobin, white blood cell count, creatinine, and alanine aminotransferase (ALT). These basic physical examination indicators are core quantitative indicators reflecting the basic health status of the human body and related to the risk of various diseases, used for calculating disease risk probabilities.

[0008] Preferably, the personal health basic profile set consists of several independent personal health basic profiles. Each personal health basic profile uses the user ID as a unique primary key. Through this primary key, the user's physical examination data generated at different times and different hospitals can be accurately associated and aggregated. The personal health basic profile associates and stores the user's previous physical examination results, as well as the basic physical examination indicators and examination time corresponding to each physical examination result, through the primary key user ID.

[0009] Preferably, the method for constructing a population disease factor model library is as follows: S1. From all the personal health basic records contained in the personal health basic record set, data is filtered and aggregated according to the type of physical examination results. The physical examination basic indicators corresponding to all personal health basic records with the same physical examination results are selected and integrated into an independent physical examination result physical examination basic indicator set. Finally, n+1 physical examination result physical examination basic indicator sets are formed. The number n+1 corresponds to all types of physical examination results, where n represents n different disease types (n≥1), and the extra 1 corresponds to health, ensuring that all physical examination results can be accurately grouped. S2. For each set of basic physical examination indicators with the same physical examination results obtained in S1, dynamic density segmentation processing is used to finally obtain the optimal distribution range set of physical examination indicators corresponding to the set of basic physical examination indicators with the same physical examination results. The optimal distribution range set of physical examination indicators contains the optimal distribution range of 10 independent indicator elements, which correspond to the 10 indicator elements in the basic physical examination indicators, namely BMI, fasting blood glucose, systolic blood pressure, diastolic blood pressure, total cholesterol, triglycerides, hemoglobin, white blood cell count, creatinine and alanine aminotransferase. Dynamic density segmentation processing captures the concentrated distribution range of the values ​​of each indicator element in different physical examination result groups by statistically analyzing the distribution range density characteristics of the indicator element values, ensuring that the optimal distribution range of each indicator element can truly reflect the corresponding physical examination result group. S3. Associate and bind the physical examination results corresponding to the basic physical examination index set of each identical physical examination result in S1 with the optimal distribution range set of physical examination indicators obtained in S2 to form a one-to-one mapping relationship. The physical examination results of n disease types are associated with their respective optimal distribution range sets of physical examination indicators to form an n-disease index mapping model. The physical examination results of healthy status are associated with their corresponding optimal distribution range sets of physical examination indicators to form a health index mapping model. The n-disease index mapping model and the health index mapping model are aggregated together to form a group disease factor model library.

[0010] Preferably, a method for dynamically segmenting the basic indicators for each identical physical examination result is used: S01. Statistically analyze the range of indicator values ​​and the amount of indicator data corresponding to the indicator elements in the set of basic indicators for physical examinations with the same results. The range of indicator values ​​is the interval between the minimum and maximum values ​​of the optimal distribution range of the indicator elements in the set of optimal distribution ranges of the physical examination indicators for that indicator element. The amount of indicator data is the amount of data in the optimal distribution range of the indicator elements in the set of optimal distribution ranges of the physical examination indicators for that indicator element. Based on the range of indicator values ​​and the amount of indicator data corresponding to the indicator element, calculate the initial segment length of the indicator element using the initial segment length formula, and divide the range of indicator values ​​of the indicator element into k consecutive sub-range segments according to the initial segment length. The formula for the initial segment length is: ; in, It refers to the amount of indicator data. It is the upper limit of the range of indicator values. It is the upper limit of the range of indicator values. It is the initial segment length; S02. Calculate the amount of data within each sub-range of the indicator value, i.e., the amount of indicator data falling into that sub-range, and optimize for two special cases: If there is a sub-range of index values ​​with 0 data points, it means that there is no data distribution within that sub-range of index values. In this case, the sub-range of index values ​​will be merged with the sub-range of index values ​​on the adjacent side. If the proportion of data within a sub-range of an index value to the total index data is greater than 20%, it indicates that the data distribution within that sub-range is too concentrated. In this case, the sub-range of an index value is divided into segments according to half the length of the sub-range. After this process, m index value segments (m≥k) are obtained. S03. Calculate the atomic segment density of each indicator value atomic segment using the density formula to reflect the degree of data concentration within the indicator value atomic segment. Select all indicator value atomic segments with atomic segment density not less than 0 to form a candidate high-density indicator value atomic segment set. If the data volume of the candidate high-density indicator value atomic segment set accounts for not less than 90% of the indicator data volume, then merge the adjacent indicator value atomic segments in the candidate high-density indicator value atomic segment set to obtain the optimal distribution range of indicator elements composed of w optimal distribution sub-ranges of indicator elements, where w is the number of merged indicator value atomic segments. This optimal distribution range of indicator elements can accurately reflect the value distribution characteristics of the indicator element in the corresponding physical examination result group. The density formula is: ; in, It is the atomic segment density. It is the original atomic segment density of the index value atomic segment. It is the density mean. It is the density standard deviation. It refers to the amount of data contained in each index value's atomic segment. It is the segment length of each index value's atomic segment; The and It is obtained by calculating the mean and standard deviation of the original atomic segment density of all index value atomic segments.

[0011] S04. If the data volume of the candidate high-density index value atomic segment set accounts for less than 90% of the index data volume, then reduce the value of the density standard deviation in the low-density formula by 0.2. By relaxing the density standard deviation threshold, the screening range of index value atomic segments is expanded, and step S03 is executed again. The indicator elements are BMI, fasting blood glucose, systolic blood pressure, diastolic blood pressure, total cholesterol, triglycerides, hemoglobin, white blood cell count, creatinine or alanine aminotransferase. The above dynamic density segmentation processing is performed independently for each different indicator element, and finally the optimal distribution range of the indicator element is generated for each different indicator element under different physical examination results, laying a quantitative foundation for the similarity analysis of the user's basic physical examination indicators and the indicator mapping model.

[0012] Preferably, the process of performing similarity analysis between the user's basic physical examination indicators and the health indicator mapping models and various disease indicator mapping models in the group disease factor model library is as follows: The user's basic physical examination indicators, including BMI, fasting blood glucose, systolic blood pressure, diastolic blood pressure, total cholesterol, triglycerides, hemoglobin, white blood cell count, creatinine, and alanine aminotransferase (ALT), are matched with the optimal distribution range sets of physical examination indicators corresponding to the health indicator mapping model and various disease indicator mapping models in the population disease factor model library. For each user's basic physical examination indicator, its specific value is located within the optimal distribution sub-range of the optimal distribution range set of physical examination indicators under the health indicator mapping model or various disease indicator mapping models. This optimal distribution sub-range is the optimal distribution sub-range of one of the optimal distribution sub-ranges of w indicator elements. This forms the optimal distribution sub-range set of user's basic physical examination indicators under the health indicator mapping model, and the optimal distribution sub-range set of user's basic physical examination indicators under various disease indicator mapping models. The optimal distribution sub-range set of indicators includes the optimal distribution sub-ranges of BMI, fasting blood glucose, systolic blood pressure, diastolic blood pressure, total cholesterol, triglycerides, hemoglobin, white blood cell count, creatinine, and ALT. The weights of each user's basic health checkup indicator are calculated based on the optimal distribution sub-range set to which the indicator belongs. These include the weights of BMI, fasting blood glucose, systolic blood pressure, diastolic blood pressure, total cholesterol, triglycerides, hemoglobin, white blood cell count, creatinine, and alanine aminotransferase (ALT). For each basic health checkup indicator, its weight is the proportion of the data volume of the optimal distribution sub-range to which the corresponding indicator belongs to the total data volume of the optimal distribution range of the corresponding indicator element under the health indicator mapping model or disease indicator mapping model. The calculation formula is as follows: ; in, It is the weight of the i-th basic health check indicator for users. It is the amount of data in the optimal distribution subrange corresponding to the i-th basic user health check indicator. It is the total amount of data for the optimal distribution range of the indicator elements corresponding to the i-th basic physical examination indicator under the health indicator mapping model or disease indicator mapping model, i=1,2,...,10, corresponding to BMI, fasting blood glucose, systolic blood pressure, diastolic blood pressure, total cholesterol, triglycerides, hemoglobin, white blood cell count, creatinine and alanine aminotransferase respectively; By substituting the weights of various basic health checkup indicators into the disease risk probability formula, the health probability and various disease risk probabilities corresponding to the basic health checkup indicators are calculated, and finally a personal disease risk probability sequence containing all disease risk probabilities and health probabilities is formed. The formula for the probability of disease risk is: ; in, This represents the probability of being healthy or the probability of risking a certain disease, and its value ranges from 0 to 1. It is the weight of the i-th basic indicator of user health check.

[0013] Preferably, the risk level of the individual's disease risk probability sequence is judged. If the health probability is greater than 0.98 and the risk probabilities of each disease are not greater than 0.4, then the user's physical examination disease set is empty, which is used to represent the health of the physical examination user; otherwise, the disease types corresponding to all disease risk probabilities greater than 0.4 are combined to form the user's physical examination disease set.

[0014] Preferably, the user's set of physical examination diseases is matched with a physical examination screening program through a disease physical examination rule engine. At the same time, if the user's set of physical examination diseases is empty, the physical examination screening program will directly state that no further physical examination screening is required. The disease examination rule engine contains a mapping relationship between disease types and examination screening program schemes.

[0015] Due to the adoption of the above technical solution, the technical progress achieved by this invention compared to the prior art is as follows: 1. This invention improves the scientific rigor and accuracy of disease risk assessment. Traditional techniques rely on physician experience or fixed thresholds for risk assessment, failing to fully utilize population data to build quantitative models and easily leading to misjudgments due to individual differences. This invention, based on an individual's basic health record set, groups indicators according to physical examination results to form an indicator set. Then, through dynamic density segmentation processing, it mines the optimal distribution range of each indicator in different groups, accurately capturing the true distribution characteristics of indicators in specific health or disease states. Based on this, it constructs a population disease factor model library containing n disease indicator mapping models and 1 health indicator mapping model, breaking through the limitations of fixed thresholds and providing a quantitative benchmark based on large-scale population data for individual risk assessment, upgrading risk assessment from experience-based judgment to data-driven precise quantification.

[0016] 2. This invention enables personalized disease risk assessment, quantifies risk probabilities, and accurately identifies potential health risks. Existing risk assessment technologies often rely on comparisons of single indicators or qualitative judgments, which are difficult to reflect the correlation between indicators and the differences between individuals and groups. This invention analyzes the similarity between the user's basic physical examination indicators and the mapping model in the group disease factor model library to form a quantified personal disease risk probability sequence. By determining the risk level through clear thresholds, it accurately identifies high-risk disease types, greatly improving the scientific nature and accuracy of disease risk assessment, and can effectively identify potential disease risks.

[0017] 3. This invention improves the efficiency of physical examinations and the effectiveness of health management. By using the user's set of diseases for physical examinations, it calls a disease examination rule engine that includes a mapping relationship between disease types and screening items, and matches targeted physical examination screening plans. This breaks away from the one-size-fits-all approach to physical examination items, reducing unnecessary physical examination items for low-risk groups while ensuring that high-risk groups receive targeted screening, thus significantly improving the efficiency of physical examinations and the accuracy of health management. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0019] Figure 1 This is a schematic diagram of the system functional modules of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Examples, such as Figure 1 The aforementioned health check-up system based on medical data analysis includes a data acquisition module, a data integration module, an intelligent analysis module, and a medical module; The data acquisition module is used to acquire all users' physical examination data; The data integration module is used to construct a set of basic personal health records based on all users' physical examination data and store them in the database, and to construct a group disease factor model library based on the set of basic personal health records. The intelligent analysis module is used to perform similarity analysis between the user's basic physical examination indicators and the health indicator mapping model and various disease indicator mapping models in the group disease factor model library to obtain an individual disease risk probability sequence, and to judge the risk level of the individual disease risk probability sequence to obtain the user's physical examination disease set. The medical module is used by doctors to push the selected user's basic physical examination indicators to the intelligent analysis module. At the same time, it matches the physical examination screening program plan based on the user's physical examination disease set, and pushes the individual disease risk probability sequence and physical examination screening program plan to the physical examination user to which the user's basic physical examination indicators belong via SMS.

[0022] Furthermore, the working principle of the present invention will be illustrated below through embodiments: This system is deployed on a local server in a hospital and connects to the hospital's information system and laboratory information system via the hospital's local area network. It supports the processing capacity of physical examination data for an average of 500 people per day. The database of this system adopts a MySQL cluster to ensure data storage security and efficient access. At the same time, the intelligent analysis module is equipped with a Python data analysis engine, which supports parallel computing for dynamic density segmentation processing and risk probability calculation.

[0023] The data acquisition module connects to the hospital's HIS and LIS in real time via the HL7FHIR protocol, automatically capturing user physical examination data including user ID, examination time, examination results, and basic physical examination indicators. The user ID is generated by encrypting the user's ID number using the SHA-256 algorithm. For example, if the user's ID number is 110101XXXX01011234, the encrypted result would be "a7b3f9d2e4c8a1b2c3d4e5f6a7b8c9d0e1f2a". The examination time is October 15, 2023. The examination results are determined by the intelligent analysis module and are currently "pending analysis". Basic physical examination indicators include a BMI of 23.5 kg / m². 2 The patient's fasting blood glucose was 5.2 mmol / L, systolic blood pressure was 125 mmHg, diastolic blood pressure was 80 mmHg, total cholesterol was 4.8 mmol / L, triglycerides were 1.6 mmol / L, hemoglobin was 145 g / L, and white blood cell count was 6.2 × 10⁻⁶. 9 / L, creatinine was 85μmol / L and alanine aminotransferase was 25U / L; the HIS stores the basic information of the physical examination user and the examination time, while the LIS stores the basic physical examination indicators and examination results of the physical examination user.

[0024] The data integration module constructs a personal health profile for the user based on their physical examination data. Using the user ID as the unique primary key, it links the user's physical examination data from the past three years to form a personal health profile. This profile records data such as the user's physical examination data from 2021, 2022, and the current physical examination in 2023. Corresponding personal health profiles are then constructed for other users at the hospital, forming a set of personal health profiles. Assuming this set contains 50,000 historical physical examination records from the entire hospital, including data from healthy individuals and patients with n=5 diseases such as hypertension and diabetes, the following operations are performed: Based on the physical examination results, 50,000 historical physical examination data were divided into 6 sets of basic physical examination indicators with the same physical examination results. For example, the set of basic physical examination indicators for hypertension contains 12,000 historical physical examination data with systolic blood pressure ≥140 mmHg, and the set of basic physical examination indicators for health contains 30,000 historical physical examination data. Taking systolic blood pressure as an example from the basic health checkup indicator set, the statistical range of systolic blood pressure values ​​is 80-130 mmHg, with a data volume of 30,000 records. The initial segment length is calculated to be approximately 0.29 mmHg using the initial segment length formula. Based on this, the range of systolic blood pressure values ​​is divided into 172 sub-ranges. Empty segments are merged, such as the systolic blood pressure value sub-range corresponding to 120-120.29 mmHg, which has 0 data records. High-density segments are further subdivided, such as the systolic blood pressure value sub-range corresponding to 110-110.29 mmHg, where the data volume accounts for 25% of the total indicator data. Subdivided by a segment length of 0.145 mmHg, a final total of 210 systolic blood pressure value atomic segments are obtained. The atomic segment density of each systolic blood pressure index value is calculated. All atomic segments of systolic blood pressure index values ​​with an atomic segment density ≥ 0 are selected and merged to obtain the optimal distribution range set of systolic blood pressure for healthy individuals. This optimal distribution range set of systolic blood pressure for healthy individuals covers 92% of the index values ​​related to systolic blood pressure in the basic health checkup index set. Similarly, the same processing is applied to BMI, fasting blood glucose, diastolic blood pressure, total cholesterol, triglycerides, hemoglobin, white blood cell count, creatinine, and alanine aminotransferase in the basic health checkup index set as to systolic blood pressure, resulting in the optimal distribution range set of 9 index elements, including the optimal distribution range set of BMI and the optimal distribution range set of fasting blood glucose for healthy individuals. The health checkup result of "healthy" is associated with the optimal distribution range set of 10 index elements to form a health index mapping model. By applying the same method to five sets of basic health examination indicators with the same results, such as the basic health examination indicator set for hypertension and the basic health examination indicator set for diabetes, we obtained five disease indicator mapping models, which were then aggregated into a population disease factor model library.

[0025] The intelligent analysis module matches the basic physical examination indicators from the 2023 medical examination in the individual's basic health record corresponding to a7b3f9d2e4c8a1b2c3d4e5f6a7b8c9d0e1f2a, i.e., the user's basic physical examination indicators, with the group disease factor model library. Taking the matching of the health indicator mapping model as an example, if the user's basic physical examination indicator of systolic blood pressure of 125 mmHg falls within a certain sub-range of the optimal distribution range of systolic blood pressure in a healthy population, and the user's basic physical examination indicator of fasting blood glucose falls within a certain sub-range of the optimal distribution range of fasting blood glucose in a healthy population, then the optimal distribution sub-range set of the user's basic physical examination indicators under the health indicator mapping model is obtained. The weight of each user's basic physical examination indicator is calculated, such as systolic blood pressure weight of 0.8, fasting blood glucose weight of 0.9, etc. According to the disease risk probability formula, the health probability of the user's basic physical examination indicators is calculated to be 0.82.

[0026] Based on the same operation of matching the user's basic physical examination indicators with the group disease factor model library, the probability of hypertension risk for this user's basic physical examination indicators is 0.65, and the probability of risk for the other 4 diseases is less than 0.4. Since the probability of health is less than 0.98 and the probability of hypertension risk is greater than 0.4, the user's physical examination disease set is hypertension.

[0027] After reviewing the user's medical examination disease set through the medical module, the doctor calls the disease examination rule engine to match the corresponding medical examination screening program for hypertension, such as 24-hour ambulatory blood pressure monitoring and electrocardiogram examination. The doctor then automatically pushes the user's personal disease risk probability sequence and medical examination screening program to the user's mobile phone via SMS, with the message: "Your 2023 medical examination risk assessment shows: hypertension risk 65%, it is recommended to undergo 24-hour ambulatory blood pressure monitoring and electrocardiogram examination."

[0028] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A health check-up system based on medical data analysis, characterized in that, include: The data acquisition module is used to acquire all users' physical examination data; The data integration module is used to build a set of basic personal health records based on all users' physical examination data and store them in the database, and to build a group disease factor model library based on the set of basic personal health records. The intelligent analysis module is used to perform similarity analysis between the user's basic physical examination indicators and the health indicator mapping models and various disease indicator mapping models in the group disease factor model library to obtain the individual disease risk probability sequence, and to judge the risk level of the individual disease risk probability sequence to obtain the user's physical examination disease set. The medical module allows doctors to push selected user health checkup indicators to the intelligent analysis module. At the same time, it matches health checkup screening programs based on the user's set of health checkup diseases and pushes the individual's disease risk probability sequence and health checkup screening program program to the user whose health checkup indicators belong via SMS.

2. The health check-up system based on medical data analysis according to claim 1, characterized in that, The user's physical examination data is obtained by connecting to the hospital's HIS and LIS via the HL7FHIR protocol. Specifically, it includes the user ID, examination time, examination results, and basic physical examination indicators. The basic physical examination indicators include BMI, fasting blood glucose, systolic blood pressure, diastolic blood pressure, total cholesterol, triglycerides, hemoglobin, white blood cell count, creatinine, and alanine aminotransferase. The physical examination results are either healthy or n diseases, where n is at least 1, and n diseases represent n different disease types.

3. The health check-up system based on medical data analysis according to claim 2, characterized in that, The user ID is obtained by irreversibly encrypting the medical insurance electronic voucher code, electronic health card number, or ID card number using SHA-256.

4. The health check-up system based on medical data analysis according to claim 3, characterized in that, The personal health basic profile set consists of several personal health basic profiles. Each personal health basic profile uses the user ID as the primary key and is associated with each physical examination result, as well as the corresponding basic physical examination indicators and examination time.

5. The health check-up system based on medical data analysis according to claim 4, characterized in that, The method for constructing a population disease factor model library: S1. Select the basic physical examination indicators corresponding to the same physical examination results from all the basic physical examination records in the personal health basic record set, and form n+1 sets of basic physical examination indicators with the same physical examination results. Each set of basic physical examination indicators with the same physical examination results in the n+1 sets of basic physical examination indicators with the same physical examination results corresponds to a different physical examination result. S2. Perform dynamic density segmentation processing on each set of basic physical examination indicators with the same physical examination results to obtain the set of optimal distribution ranges of physical examination indicators corresponding to the set of basic physical examination indicators with the same physical examination results. The set of optimal distribution ranges of physical examination indicators consists of the optimal distribution ranges of 10 indicator elements, and each optimal distribution range of indicator element corresponds to a different indicator element. S3. Associate the physical examination results corresponding to each set of basic physical examination indicators with the optimal distribution range set of the corresponding physical examination indicators for each set of identical physical examination results, to obtain n disease indicator mapping models and 1 health indicator mapping model, and thus construct a group disease factor model library including n disease indicator mapping models and 1 health indicator mapping model.

6. The health check-up system based on medical data analysis according to claim 5, characterized in that, The method for dynamically density-segmenting each set of basic physical examination indicators with the same physical examination results: S01. Statistically analyze the range of indicator values ​​and the amount of indicator data corresponding to the indicator elements in the basic indicator set for the same physical examination results. Based on the range of indicator values ​​and the amount of indicator data, calculate the initial segment length of the indicator element using the initial segment length formula, and divide the range of indicator values ​​corresponding to the indicator element into segments according to the initial segment length to obtain k consecutive sub-range segments of indicator values. S02. Count the amount of data within each sub-range of the indicator value. If there is a sub-range of the indicator value with a data amount of 0, merge the sub-range of the indicator value with the sub-range of the indicator value on the adjacent side of the sub-range of the indicator value. If the proportion of data within a sub-range of index values ​​to the total index data is greater than 20%, then the sub-range of index values ​​is divided into segments according to half the length of the sub-range of index values, thereby obtaining m atomic segments of index values. S03. Calculate the atomic segment density of each index value atomic segment using the density formula, select all index value atomic segments with atomic segment density not less than 0 to form a candidate high-density index value atomic segment set. If the data volume of the candidate high-density index value atomic segment set accounts for not less than 90% of the index data volume, then merge the adjacent index value atomic segments in the candidate high-density index value atomic segment set to obtain the optimal distribution range of index elements composed of w optimal distribution sub-ranges of index elements. S04. If the data volume of the candidate high-density index value atomic segment set accounts for less than 90% of the index data volume, then reduce the value of the density standard deviation in the low-density formula by 0.2 and repeat S03. The indicators are BMI, fasting blood glucose, systolic blood pressure, diastolic blood pressure, total cholesterol, triglycerides, hemoglobin, white blood cell count, creatinine or alanine aminotransferase.

7. The health check-up system based on medical data analysis according to claim 6, characterized in that, The process of performing similarity analysis between the user's basic physical examination indicators and the health indicator mapping models and various disease indicator mapping models in the group disease factor model library: The user's basic physical examination indicators, including BMI, fasting blood glucose, systolic blood pressure, diastolic blood pressure, total cholesterol, triglycerides, hemoglobin, white blood cell count, creatinine, and alanine aminotransferase, are paired with the optimal distribution ranges of 10 indicator elements in the optimal distribution range set of physical examination indicators in the health indicator mapping model and various disease indicator mapping models. This yields the optimal distribution sub-range set of the user's basic physical examination indicators under the health indicator mapping model, and the optimal distribution sub-range set of the user's basic physical examination indicators under various disease indicator mapping models. The optimal distribution sub-range set of the indicators includes the optimal distribution sub-ranges of BMI, fasting blood glucose, systolic blood pressure, diastolic blood pressure, total cholesterol, triglycerides, hemoglobin, white blood cell count, creatinine, and alanine aminotransferase. Based on the optimal distribution sub-range set to which the indicators belong, the weights of BMI, fasting blood glucose, systolic blood pressure, diastolic blood pressure, total cholesterol, triglycerides, hemoglobin, white blood cell count, creatinine, and alanine aminotransferase are obtained. These weights are then used in the disease risk probability formula to calculate the health probability and various disease risk probabilities corresponding to the user's basic physical examination indicators, forming a personal disease risk probability sequence that includes various disease risk probabilities and health probabilities.

8. The health check-up system based on medical data analysis according to claim 7, characterized in that, The risk level judgment is performed on the individual disease risk probability sequence. If the health probability is greater than 0.98 and the risk probabilities of various diseases are not greater than 0.4, then the user's physical examination disease set is empty, which is used to represent that the physical examination user is healthy; otherwise, the disease types corresponding to all disease risk probabilities greater than 0.4 are combined to form the user's physical examination disease set.

9. A health check-up system based on medical data analysis according to claim 8, characterized in that, The user's medical examination disease set is matched with the medical examination screening project plan through the disease medical examination rule engine. At the same time, if the user's medical examination disease set is empty, the medical examination screening project plan is directly set to no further medical examination screening is required. The disease examination rule engine contains a mapping relationship between disease types and examination screening program schemes.

Images