Health care information network

Abstract

The present invention establishes a network that enables easy access to results of estimation of risk for various diseases.　Provided is a health care information network comprising an information provision-side system that is provided with: a basic information table in which is registered basic data including a personal ID of a subject and health care data of the subject; an analysis engine which has plurality of estimation models optimized per disease by machine learning, and which estimates the degree of disease risk per disease upon input of the basic data into the estimation models; and a master database in which is registered degree of disease risk data relating to the degree of disease risk per disease, the basic data, and the personal ID.

Description

Healthcare Information Network

[0001] The present invention relates to healthcare information networks and the like.

[0002] In recent years, AI has been developed that can accurately estimate the risk of various diseases.

[0003] For example, Patent Document 1 discloses a method for determining the risk of liver diseases such as fatty liver, cirrhosis, and liver cancer by using attribute data such as gender and age, physical examination data such as height and weight, and blood test data such as AST and ALT.

[0004] Patent No. 7170368

[0005] According to the technique disclosed in Patent Document 1, the risk of liver disease and the like can be estimated with high accuracy and ease by utilizing machine learning.

[0006] These assessment methods using machine learning are being expanded to assess not only the risk of liver disease, but also the risk of various diseases such as brain disease and diabetes. It is hoped that a network environment that allows easy access to such information will be established soon in order to predict and prevent various diseases and ultimately improve people's health.

[0007] The present invention has been made in view of the above-mentioned problems, and aims to provide a network that allows easy access to the results of estimation of various disease risks.

[0008] The present invention provides a healthcare information network including: an information providing system including: a basic information table in which basic data including a subject's personal ID and the subject's healthcare data is recorded; an analysis engine having a plurality of estimation models optimized for each disease type by machine learning, which estimates a disease risk level for each disease type by inputting the basic data into the estimation model; and a master database in which the personal ID, the basic data, and the disease risk level for each disease type are registered; and a user terminal that enables a user who wishes to use the disease risk levels to access the master database of the information providing system.

[0009] According to the present invention, a network can be constructed that allows various institutions to easily access the results of estimation of various disease risks, making it possible to provide high-quality healthcare services.

[0010] Schematic diagram showing the overall configuration of a healthcare information network according to one embodiment of the present invention. Schematic diagram showing the main functions of a healthcare information network according to one embodiment of the present invention. Table showing an outline of an estimation model for each disease type used in one embodiment of the present invention. Table showing an outline of a first database (A) and a second database (B) used in one embodiment of the present invention. Schematic diagram for explaining the outline of Application Example 1. Flow diagram for explaining the outline of Application Example 1. Schematic diagram for explaining the outline of Application Example 2. Flow diagram for explaining the outline of Application Example 2. Schematic diagram for explaining the outline of Application Example 3. Schematic diagram for explaining the outline of Application Example 4. Schematic diagram for explaining the outline of Application Example 5.

[0011] The following describes a healthcare information network 10, an information provider system 20 (system), and a method for acquiring disease risk data according to one embodiment of the present invention. The healthcare information network 10 includes an information provider-side information processing device (hereinafter sometimes referred to as the information provider system 20), which is made up of institutions, corporations, individuals, etc. that provide risk level information related to the risk levels of various diseases, and a user-side information processing device (hereinafter sometimes referred to as the user terminal 100), which is made up of institutions, corporations, individuals, etc. that wish to use the risk level information related to the risk levels.

[0012] The information providing system 20 includes a basic information table 30, an analysis engine 40, and a master database 50. The basic information table 30 is a table in which basic data including the subject's personal ID and the subject's healthcare data is recorded. The analysis engine 40 has a plurality of disease risk estimation models optimized for each disease type by machine learning, and estimates the disease risk for each disease type by inputting the basic data into the estimation model. The master database 50 is a table in which the personal ID, the basic data, and disease risk data relating to the disease risk for each disease type are registered.

[0013] The user terminal 100 is provided by a user institution (user) that wishes to use disease risk data, such as a clinic, medical institution, insurance company, fitness gym, drug discovery company, research institute, etc., and can be used by the user to access the master database 50 of the information provider system 20.

[0014] The healthcare information network 10 allows users at each user institution to easily access highly accurate estimates of various disease risks provided by information providers. Such highly accurate and easily accessible healthcare services lead to improved health for subjects and more effective use of medical resources.

[0015] In the healthcare information network 10 of this embodiment, the master database 50 has a first database 52 and a second database 54. The first database 52 is a database in which the disease risk data and basic data corresponding to individual IDs are registered. The second database 54 is a database in which the disease risk data and basic data corresponding to disease types are registered. The healthcare information network 10 can set access rights to either or both of the first database 52 and the second database 54 for each user.

[0016] Specifically, when a user is an institution such as a clinic, insurance company, or fitness gym, the healthcare information network 10 can set access rights for the user to the first database 52 in which disease risk data and basic data linked to personal IDs are registered, and when a user is a drug discovery company, research institution, or the like, the healthcare information network 10 can set access rights for the user to the second database 54 in which disease risk data and basic data linked to disease types, i.e., not linked to personal IDs. Note that the healthcare information network 10 can also limit the types of basic data that can be extracted from the first database 52 in response to requests from users such as institution users.

[0017] In this way, the healthcare information network 10 can preliminarily separate accessible databases according to the user, thereby increasing the safety of information and improving security.

[0018] In the healthcare information network 10 of this embodiment, the basic data preferably includes the subject's health checkup / examination data or the subject's daily monitoring data. Using both the health checkup / examination data and the daily monitoring data may further improve the accuracy of the risk levels of various diseases.

[0019] Examples of health checkup and examination data include attribute data, physical examination data, specimen test data obtained from blood tests and urine tests, and physiological function test data such as ECG. These data are one-time monitoring data. On the other hand, daily monitoring data is data obtained from general-purpose wearable devices such as smart watches, medical devices, and medical terminals, and includes, for example, blood glucose levels, ECG, SPO, etc. ２ Examples of such data include continuous monitoring data of blood pressure, body temperature, etc. By using both one-time monitoring data and continuous monitoring data, the risk of various diseases can be evaluated with higher accuracy.

[0020] Attribute data include gender, age, medical history, presence or absence of underlying diseases, race, genetic predisposition (e.g., PNPLA3), etc. Physical examination data include height, weight, blood pressure, waist circumference, vision, hearing, body fat percentage, sitting height, head circumference, chest circumference, etc. Laboratory test data include AST (also known as GOT), ALT (also known as GPT), γ-GTP, PLT, T-Cho, TG, Alb, HDL, LDL, HbA1c, ALP, ChE, type 4 collagen (especially type IV collagen 7S), Total-AIM, Free-AIM, etc. Physiological function test data include hardness and elastography using ultrasound and magnetic resonance, X-ray images, CT images, MRI images, endoscopic images of various parts including the chest, and interview results. Other basic data may include interview data obtained through remote interviews, etc.

[0021] In the healthcare information network 10, the analysis engine 40 preferably includes a first engine that estimates a disease risk level based on basic data, and a second engine that estimates the disease risk level based on either the basic data or the analysis results by the first engine. With this configuration, primary screening for assisting in the diagnosis of each disease can be performed by the first engine, and secondary screening for future prediction of each disease can be performed by the second engine.

[0022] In addition, in the healthcare information network 10, the analysis engine 40 may further include a third engine that estimates the disease risk level based on at least one of the basic data, the analysis results by the first engine, and the analysis results by the second engine. This configuration enables more accurate diagnosis of each disease.

[0023] The present invention is also directed to a system for providing risk levels for various diseases. An information providing system 20, which is one embodiment of this system, includes a basic information table 30 in which basic data including the subject's personal ID and healthcare data is recorded, an analysis engine 40 having a plurality of estimation models optimized for each disease type by machine learning and estimating the disease risk level for each disease type by inputting the basic data into the estimation models, and a master database 50 in which the personal ID, the basic data, and disease risk level data relating to the disease risk level for each disease type are registered.

[0024] This system, that is, the information provider system 20 which is an information processing device on the information provider side, can highly accurately evaluate the risk levels of multiple diseases based on limited basic data.

[0025] The present invention is also directed to a method for acquiring disease risk data for various diseases. This method for acquiring disease risk data can be realized, for example, in the above-described healthcare information network 10 and information provider system 20. Specifically, the method for acquiring disease risk data of the present invention includes the steps of inputting basic data including a subject's personal ID and healthcare data, inputting the basic data into an analysis engine 40 having multiple estimation models optimized for each disease type through machine learning to estimate a disease risk for each disease type, registering the personal ID, the basic data, and disease risk data relating to the disease risk for each disease type in a master database 50, and accessing the master database 50 to obtain the disease risk data. The method for acquiring disease risk data including these steps can be executed by the above-described healthcare information network 10 and information provider system 20.

[0026] This method of acquiring disease risk data can be performed in a virtual space (metaverse). That is, the subject visits the virtual space as an avatar, visits desired information use institutions (users) within the virtual space, provides healthcare information including the basic data to the information users, and receives necessary information from each user. The user can provide basic data to an information provider outside the virtual space, obtain disease risk data from the information provider, and use the data. Note that the above-described method of acquiring disease risk data can also be performed in a manner in which the subject provides basic data to an information provider established within the virtual space, and the information provider evaluates various disease risks within the virtual space.

[0027] The present invention also relates to a healthcare information network 10 including an information provider system 20 including a basic information table 30 that records basic data including a subject's personal ID and the subject's healthcare data; an analysis engine 40 that has an estimation model optimized by machine learning and that estimates a disease risk level for a specific disease type by inputting the basic data into the estimation model; and a master database 50 that registers disease risk level data related to the personal ID, the basic data, and the disease risk level; and a user terminal 100 provided to a user who wishes to use the disease risk level data and that can access the master database 50 of the information provider system 20. This healthcare information network 10 allows each user to easily access highly accurate disease risk level estimation results provided by an information provider. Such highly accurate and easily accessible healthcare services lead to improved health for subjects and more effective use of medical resources.

[0028] The present invention is also directed to an analysis engine 40 that has multiple estimation models optimized for each disease type by machine learning, and estimates a disease risk for each disease type by inputting basic data including the subject's healthcare data into the estimation models. Such analysis engine 40 can accurately estimate a disease risk for each disease type based on the basic data including the subject's healthcare data.

[0029] The present invention will be described below in accordance with the above-described embodiment, illustrating specific examples of the healthcare information network 10, the information provider system 20, and the method for acquiring disease risk data, but the present invention is not limited to these examples. That is, the healthcare information network, the information provider system, and the method for acquiring disease risk data of the present invention can be other than the examples illustrated below, as long as they do not deviate from the spirit of the present invention.

[0030] A preferred embodiment of a healthcare information network 10 or the like will now be described with reference to FIGS.

[0031] As shown in FIG. 1, a healthcare information network 10 of this embodiment is an IT network system between an information processing device on the information providing side and users who use the information processing device.

[0032] The information processing device on the information provider side, that is, the information provider system 20, has an analysis database (analysis DB) and a master database 50 (master DB).

[0033] The analysis database includes a basic information table 30 in which the subject's personal ID and basic data related to healthcare of the subject are recorded, an analysis engine 40 having multiple estimation models optimized for each disease type through machine learning, and a result information table 60 in which disease risk data for each type of disease predicted and estimated by the analysis engine 40 is registered.

[0034] The master database 50 stores personal IDs, basic data (healthcare information data), and disease risk data for each disease type. The master database 50 includes a first database 52 (DB-1) and a second database 54 (DB-2). The first database 52 stores personal IDs, basic data corresponding to these personal IDs, and disease risk data for each disease type. The second database 54 stores basic data corresponding to each disease type and the risk of the disease.

[0035] Each user institution (user), such as a clinic (medical institution), insurance company, fitness gym, drug discovery company, or research institute, has an information processing device (user terminal 100). This system can access a master database 50 of the information providing system 20. In other words, when the user terminal 100 requests the information providing system 20 to provide necessary information, the information providing system 20 provides the user terminal 100 with necessary information such as basic data on healthcare-related information and the disease risk level for each type of disease.

[0036] The basic information table 30 in the information providing system 20 receives basic data, i.e., healthcare-related information, such as one-time monitoring data obtained from health checkups and examinations, and continuous monitoring data obtained from wearable devices, medical equipment, medical terminals, etc.

[0037] The analysis engine 40 in the information providing system 20 may be, for example, a program that executes the disease risk assessment method disclosed in Japanese Patent No. 7170368 (Patent Document 1). As disclosed in Patent Document 1, the analysis engine 40 has a disease risk estimation model constructed through preprocessing of training data, neural networking, deep learning, and postprocessing. The analysis engine 40 is equipped with multiple estimation models optimized for multiple diseases. The analysis engine 40 includes a diagnostic engine (first engine) that performs primary screening to assist in the diagnosis of each disease, a prediction engine (second engine) that performs secondary screening to predict the future course of each disease, and a diagnostic engine (third engine) that assists in more accurate diagnosis of each disease, as will be described in detail below. These first, second, and third engines each include an estimation model with an optimized algorithm for estimating the disease risk of each disease.

[0038] Estimation result data of each disease risk level calculated by the analysis engine 40 is stored as disease risk level data together with the personal ID in the result information table 60. Note that this information processing device does not need to have all of the functional units described below within the information provider; for example, the analysis engine 40 and master database 50 may be provided outside the information provider.

[0039] As shown in FIG. 2 , the healthcare information network 10 of this embodiment utilizes data (one-time monitoring data) obtained through health checkups and medical examinations, as well as continuous sensing data (continuous monitoring data) obtained from wearable devices, medical devices, medical terminals, etc. The network then utilizes an analysis engine 40, which includes a diagnostic engine for diagnosing the current risk of various diseases and a prediction engine for predicting the future risk of various diseases, to predict the risk of various diseases, as well as cancer and complications, and assist in their diagnosis and treatment. If no medical treatment is required, the network assists in the creation and implementation (progress monitoring) of a recovery program at, for example, a fitness gym. This enables early detection of abnormalities, i.e., disease risks, and assists in the subject's recovery to a healthy state.

[0040] Indicators of use include physical examinations, blood tests, chest X-rays, ultrasound echoes, and CT / MRIs for health checkups and medical examinations. Non-invasive blood glucose meters, ECGs, and SPOs are also available for continuous sensing. ２ , HbA1c, blood pressure, and remote interview can be used. Hereinafter, these index data may be referred to as basic data.

[0041] Based on the above-mentioned indicators, the current health state is first quantified using the analysis engine 40. That is, the diagnostic engine (first engine) performs primary screening to assist in diagnosis. Specifically, the following conditions are quantified: liver disease conditions for diagnosing hepatitis or cirrhosis, brain disease conditions for diagnosing dementia, diabetes conditions for diagnosing diabetes, mental disease conditions for diagnosing bipolar disorder, lung disease conditions for diagnosing pneumonia, heart disease conditions for diagnosing angina pectoris or heart failure, kidney disease conditions for diagnosing pyelitis, etc. More specifically, the diagnostic engine (first engine) includes an estimation model A that extracts specific data (here, physical finding data and blood test data) corresponding to each disease from the plurality of basic data and performs screening for hepatitis and cirrhosis based on a specific algorithm a1, an estimation model B that similarly extracts specific data (here, physical finding data and blood test data) corresponding to each disease from the plurality of basic data and performs screening for dementia based on a specific algorithm b1, an estimation model C that similarly extracts specific data (here, physical finding data, blood test data, and continuous sensing data such as blood glucose level data and blood pressure data) corresponding to each disease from the plurality of basic data and performs screening for diabetes based on a specific algorithm c1, etc. In other words, by performing primary screening using the diagnostic engine that diagnoses the current risk level of various diseases, the health condition is visualized by quantifying the risk level of the desired disease (preferably the risk levels of multiple diseases), which contributes to the early detection of abnormalities.

[0042] Furthermore, the analysis engine 40 uses the basic data and the analysis results from the diagnostic engine to predict future disease risk. That is, a prediction engine (second engine) performs secondary screening for future prediction of various diseases based on the basic data and / or data calculated by the primary screening. Specifically, the disease risk for liver disease, i.e., the liver fibrosis level, is quantified using the basic data, analysis results regarding the level of hepatitis or cirrhosis, and analysis results regarding the level of diabetes. An estimation model A is provided that performs fibrosis level screening based on a specific algorithm A2. This prediction engine also includes estimation models for other diseases based on specific algorithms for each disease using the illustrated data. That is, the prediction engine (second engine) calculates the future risk of each disease, predicts the risk of each disease, and a recovery program to return to a healthy state is developed based on the predicted value, contributing to the diagnosis of whether the patient is in a healthy state or a pre-disease state and the recovery to a healthy state.

[0043] Furthermore, the analysis engine 40 is used to quantify the disease level using the basic data, the analysis results from the first engine, the analysis results from the second engine, etc. That is, in the second diagnostic engine (third engine), a tertiary screening is performed to assist in the highly accurate diagnosis of each disease based on the data calculated in the second screening, as well as the basic data and data calculated in the first screening (data from the first diagnostic engine) as needed. Specifically, the analysis results for the fibrosis level and the analysis results for diabetes prediction are used to predict the disease level for liver cancer. More specifically, this diagnostic engine includes an estimation model A that uses data calculated in the second screening for fibrosis level, data calculated in the second screening for dementia, data calculated in the second screening for diabetes, data calculated in the second screening for manic depression, and data calculated in the second screening for heart disease, as well as the basic data and data calculated in each of the first screenings as needed, to predict liver cancer based on a specific algorithm A3. Note that this diagnostic engine also includes estimation models for other diseases based on specific algorithms for predicting each disease and complication using the illustrated data. These steps not only contribute to highly accurate diagnosis of each disease and complication, but also contribute to recovery to health.

[0044] As shown in FIG. 3 , the analysis engine 40 has multiple estimation models optimized for each disease type. For example, the analysis engine 40 has a liver disease risk estimation model (estimation model A) with a calculation algorithm (algorithm a) for estimating the risk of liver disease (disease A), a brain disease risk estimation model (estimation model B) with a calculation algorithm (algorithm b) for estimating the risk of brain disease (disease B), and a diabetes risk estimation model (estimation model C) with a calculation algorithm (algorithm c) for estimating the risk of diabetes (disease C). In other words, the analysis engine 40 estimates the risk of a specific disease by inputting various healthcare-related data (i.e., basic data, analysis results of the first screening, and analysis results of the second screening) into specific estimation models. Note that while this table illustrates the types of basic data used in the first screening, specific tables are also used to estimate the risk of each disease in the second and third screenings.

[0045] The master database 50 stores personal IDs, basic data, and disease risk data relating to the disease risk for each disease type output from the analysis engine 40. The master database 50 includes a first database 52 (DB-1) and a second database 54 (DB-2). As shown in FIG. 4(A), the first database 52 stores personal IDs, basic data corresponding to the personal IDs, and disease risk data for each disease type. As shown in FIG. 4(B), the second database 54 stores basic data corresponding to each disease type and the risk of the disease.

[0046] Hereinafter, application examples in which the healthcare information network, information provider system (system), and disease risk data acquisition method according to the present invention are applied to various linkage models will be described. Note that in the following explanation, examples will be described in which various linkage models are constructed using the healthcare information network 10, information provider system 20, and disease risk data acquisition method according to the above-described embodiment and examples, but various linkage models can also be constructed using the healthcare information network 10, information provider system 20, and basic information table 30 within the scope of the present invention.

[0047] <Application Example 1> This application example relates to an application example in which the healthcare information network, information provider system (system), and disease risk data acquisition method according to the present invention are applied to various collaboration models with clinics and medical institutions, as shown in Figures 5 and 6 .

[0048] The subject inputs basic data A into the subject's terminal. The basic data A includes attribute data such as age and gender, physical examination data such as height and weight, and blood test data such as AST (GOT), ALT (GPT), γGTP, PLT, T-Cho, TG, type IV collagen (particularly type IV collagen 7S), and AIM. This basic data may be information entered by the subject into their own terminal as the results of a health check, or may be information collected from a wearable device such as a smartwatch. The subject also selects the disease type for which a risk assessment is sought. While a risk assessment may be sought for multiple types of diseases, in this example, a risk assessment for a specific disease A (for example, liver disease) is sought.

[0049] The information providing system 20 is a system installed on the information provider's side, which may be an information providing institution, corporation, individual, or the like. The information providing system 20 is composed of, for example, one or more computers, one or more storage devices, and other devices. The information providing system 20 is equipped with the analysis database and master database 50 described above. The analysis database calculates the risk level (disease risk level α1) for disease A based on the registered basic data, and registers information related to the risk level calculation result (risk level data) together with the basic data in the master database 50. The subject accesses the master database 50 from their terminal and obtains disease risk level data related to the disease risk level α1 (primary analysis result). If, after confirming the disease risk level based on the disease risk level data, they wish to be examined at a medical institution, they select the medical institution at their terminal and authorize data sharing with the medical institution. In response, the information providing system 20 authorizes the medical institution's computer (terminal) to access the subject's basic data A and the disease risk level data related to the disease risk level α1 for disease A, which have the ID.

[0050] The medical institution accesses the information providing system 20 from a computer (terminal) on the medical institution's side and obtains basic data A linked to the personal ID and disease risk data related to the disease risk level α1 for disease A. The medical institution then examines the subject while referring to the basic data A and the disease risk data related to the disease risk level α1. If necessary, a detailed examination (additional examination) such as a CT scan, MRI, specialized marker, or genetic testing (e.g., SNP of PNPLA3) is performed. The supplemental data (additional data) obtained from the detailed examination is sent to the information providing system 20, where it is again input into the analysis database as basic data A', in which the supplemental data is added to the basic data A. This analysis database then calculates disease risk data related to the disease risk level α2 (secondary analysis result). A specialist makes a definitive diagnosis and administers any necessary treatment while referring to the basic data A' and the disease risk data related to the disease risk level α2.

[0051] Although the above describes the step of estimating the risk level for the type of disease desired by the subject, it is also possible to use the system for comprehensive judgment, such as determining the degree of risk the subject has for each disease.

[0052] For example, if a subject inputs into the subject's terminal, "I would like a comprehensive assessment of the risk of each disease," the provider institution system will calculate a disease risk level α1 for disease A, as well as a disease risk level β1 for disease B, a disease risk level γ1 for disease C, and so on. In other words, a computer or the like constituting the information provider system 20 will evaluate high-risk and low-risk diseases and present this on the subject's terminal. Thereafter, as in the above, the subject may request a diagnosis from a medical institution if necessary, and additional tests may be conducted at the medical institution as necessary. The provider institution will then calculate a more accurate disease risk level (β2, γ2, ...) for each disease based on the data, and treatment will be initiated at the medical institution based on the results of these secondary analyses.

[0053] <Application Example 2> This application example relates to an application example in which the healthcare information network, information provider system (system), and disease risk data acquisition method according to the present invention are applied to various collaboration models in collaboration models with drug discovery companies and research institutions, as shown in Figures 7 and 8 .

[0054] The master database 50 of the information providing system 20 stores basic data of multiple subjects A, B, C, etc., as well as disease risk levels for each of the diseases A, B, C, etc., calculated by the analysis engine 40. Each piece of basic data includes attribute data such as age and gender, physical examination data such as height and weight, and blood test data such as AST (GOT), ALT (GPT), γGTP, PLT, T-Cho, TG, type 4 collagen (particularly type IV collagen 7S), and AIM. Furthermore, based on requests from user institutions (users) such as drug discovery companies and research institutes, a data table for each risk level of, for example, disease A is prepared in the master database 50. This table contains a set of high risk levels and a set of medium risk levels for disease A.

[0055] That is, the information providing system 20 is equipped with the above-described analysis database and master database 50, and calculates the risk of various diseases based on the basic data of each subject in the analysis database, and registers the basic data and calculation results in the master database 50. Furthermore, a data table by risk level for each disease is registered.

[0056] Drug discovery companies and research institutions develop new drugs based on basic data registered by disease type and risk level in the master database 50 of the information provider system 20, as well as disease risk level data related to disease risk levels. Specifically, real drugs and placebos are prepared according to risk levels, and these are administered to subjects according to their risk levels, and basic data (secondary data) after administration is obtained. In other words, drugs adjusted according to risk levels are provided to subjects, and basic data is periodically collected and sent to the provider institution, which then recalculates the risk level (secondary) in an analysis database and provides the results to the user institution (user). In this way, by updating analysis result indicators (disease risk level data) based on the drug administration process and observing their changes, drug efficacy can be efficiently confirmed and drug discovery can be made more efficient.

[0057] In addition, medication standards (such as drug names and dosages) based on disease risk levels are registered in a database at the provider institution, and subjects can use this database as a reference when purchasing the necessary medications at a pharmacy or drug store.

[0058] <Application Example 3> This application example relates to an application example in which the healthcare information network, information provider system (system), and disease risk data acquisition method according to the present invention are applied to various collaboration models in a collaboration model with an insurance company, as shown in FIG. 9 .

[0059] The subject provides basic data of the subject to the information providing system 20. This basic data includes attribute data such as age and sex, physical examination data such as height and weight, and blood test data such as AST (GOT), ALT (GPT), γGTP, PLT, T-Cho, TG, type 4 collagen (especially type IV collagen 7S), and AIM. The subject also designates an insurance company.

[0060] The information providing system 20 is equipped with the analysis database and master database 50 as described above, and calculates the risk of various diseases based on the subject's basic data in the analysis database, and registers the basic data and calculation results in the master database 50.

[0061] The designated insurance company accesses the provider's master database 50 from the insurance company's information terminal and obtains disease risk data related to the desired disease risk level for the desired subject. In other words, the insurance company determines whether to enroll in insurance, calculates insurance premiums, recommends insurance types, etc. for the subject based on the disease risk data related to the disease risk level registered in the master database 50 of the information provider system and linked to the personal ID. Upon receiving the subject's designation, the insurance company can access the provider's database and obtain information on the target disease risk level. Access restrictions can be imposed on the insurance company so that information on other disease risk levels cannot be viewed, and in some cases, basic data cannot be viewed either. The insurance company can provide advice and guidance on lifestyle improvements to the subject based on the specific disease risk level and adjust insurance premiums depending on the progress of such advice. This can promote the subject's health.

[0062] <Application Example 4> This application example relates to an application example in which the healthcare information network, information provider system (system), and disease risk data acquisition method according to the present invention are applied to various collaboration models in a collaboration model with a fitness gym, as shown in FIG. 10 .

[0063] The subject provides basic data about the subject to the information providing system 20. This basic data includes attribute data such as age and sex, physical examination data such as height and weight, and blood test data such as AST (GOT), ALT (GPT), γGTP, PLT, T-Cho, TG, type 4 collagen (especially type IV collagen 7S), and AIM. The subject also specifies a fitness gym.

[0064] The information providing system 20 is equipped with the analysis database and master database 50 as described above, and calculates the risk of various diseases based on the subject's basic data in the analysis database, and registers the basic data and calculation results in the master database 50.

[0065] The designated fitness gym is registered in the master database 50 of the information provider system, and an exercise menu is formulated based on basic data linked to the personal ID and disease risk data related to the disease risk level, and this is shared with the subject. By periodically checking the disease risk level, the subject can self-manage while monitoring their symptoms, for example, by losing weight and then increasing muscle mass. In addition, exercise programs (health habit improvement programs) corresponding to the disease risk level can also be registered in the master database 50 and made accessible to the subject.

[0066] <Application Example 5> As shown in FIG. 11 , this application example relates to a method for acquiring and using disease risk data related to disease risk in a virtual space, in which the healthcare information network, information provider system (system), and method for acquiring disease risk data according to the present invention are applied to various collaboration models.

[0067] The method for acquiring disease risk data related to the disease risk level described above may be performed in a virtual space (metaverse). That is, a subject can visit the virtual space as an avatar, visit a desired information using institution (user) within this virtual space, provide healthcare information to the information using institution, and receive necessary information from each information using institution. Each information using institution cooperates with an information providing institution in the real space and provides the information using institution with the risk level of each disease.

[0068] It is preferable that this virtual space reproduces a streetscape constructed in the same way as in the real space.

[0069] The present invention can be used in healthcare services aimed at promoting health and making effective use of medical resources.

Claims

1. a basic information table in which basic data including a personal ID of a subject and health care data of the subject is recorded; an analysis engine having a plurality of estimation models optimized for each disease type by machine learning, the analysis engine estimating a disease risk level for each disease type by inputting the basic data into the estimation models; a master database in which the individual ID, the basic data, and disease risk data relating to the disease risk for each disease type are registered; An information providing system comprising: a user terminal capable of accessing the master database of the information providing system; Healthcare information networks, including:

2. the master database has a first database in which the disease risk level and the basic data corresponding to the individual ID are registered, and a second database in which the disease risk level and the basic data corresponding to the disease type are registered, The healthcare information network of claim 1 , wherein access rights to either one or both of the first database and the second database are set depending on a user who wishes to use the disease risk data.

3. 3. The healthcare information network of claim 2, wherein access rights to the first database are set for users consisting of user institutions selected from the group consisting of clinics, insurance companies, and fitness gyms, and access rights to the second database are set for users consisting of user institutions selected from the group consisting of drug discovery companies and research institutes.

4. The healthcare information network of claim 1 , wherein the basic data includes medical examination and testing data of the subject or daily monitoring data of the subject.

5. The plurality of estimation models include a first engine that estimates the disease risk level based on the basic data; A second engine that estimates the disease risk level based on either the basic data or an analysis result by the first engine; 10. The healthcare information network of claim 1, each comprising:

6. The healthcare information network of claim 5 , wherein the plurality of estimation models further include a third engine that estimates the disease risk level based on at least one of the basic data, the analysis results by the first engine, and the analysis results by the second engine.

7. a basic information table in which basic data including a personal ID of a subject and health care data of the subject is recorded; an analysis engine having a plurality of estimation models optimized for each disease type by machine learning, the analysis engine estimating a disease risk for each disease type by inputting the basic data into the estimation models; a master database in which the individual ID, the basic data, and disease risk data relating to the disease risk for each disease type are registered; A system equipped with.

8. inputting basic data including a subject's personal ID and health care data; inputting the basic data into an analysis engine having a plurality of estimation models optimized for each disease type by machine learning, and estimating a disease risk level for each disease type; registering the individual ID, the basic data, and disease risk data relating to the disease risk for each disease type in a master database; accessing the master database to obtain the desired disease risk data; A method for obtaining disease risk data, including:

9. The method for acquiring disease risk data according to claim 9 , wherein the subject visits a desired institution in a virtual space and provides the basic data to the institution in the virtual space.

10. a basic information table in which basic data including a personal ID of a subject and health care data of the subject is recorded; an analysis engine having a plurality of estimation models optimized for each disease type by machine learning, the analysis engine estimating a disease risk level for a specific disease type by inputting the basic data into a desired estimation model among the plurality of estimation models; a master database in which the individual ID, the basic data, and disease risk data relating to the disease risk level are registered; An information providing system comprising: a user terminal provided to a user who wishes to use the disease risk data and capable of accessing the master database of the information providing system; Healthcare information networks, including: