Health risk evaluation support apparatus and program

Abstract

To provide a health risk evaluation support apparatus which supports health maintenance activities by an individual or an organization by transmitting information about the health risk of an evaluation object, and a program which enables each unit of the health risk evaluation support apparatus to execute.SOLUTION: In a disease risk evaluation support system in which a medical institution terminal 2 transmits eyeground information of a user to a server 1 via a network NW, the server transmits information showing the deviation of the mental disease risk of an evaluation target to the medical institution terminal via the network, and the medical institution terminal transmits information showing the deviation of the mental disease risk of the evaluation target to an individual terminal 3-1 and / or an organizational terminal 3-2 via the network, the server as a health risk evaluation support apparatus comprises: a prediction processing unit predicting the progression risk of the health risk of the user based on the time series data of the eyeground information of the user; and a communication processing unit transmitting information about the health risk of the evaluation object including the user.SELECTED DRAWING: Figure 4

Description

[Technical field]

[0001] The present invention relates to a system that uses a system to calculate a subject's risk of being in a depressive state or developing depression from fundus information, calculates the deviation of a specific individual or specific organization compared to the overall risk data (population) of a large number of subjects, visualizes the relative level of that risk, and supports measures against a high disease risk compared to other people, other departments, other companies, or other industries. [Background technology]

[0002] As medical technology becomes more advanced, computer-based diagnostic support technology is being developed. Previous diagnostic support technology was based on the dualism of "sickness" and "health," and was limited to computers assisting doctors in determining whether a patient was sick (or not healthy) based on diagnostic guidelines.

[0003] In recent years, the concept of "pre-illness," which is not a dichotomy between health and illness, has been gaining importance. "Pre-illness" is defined as "not a dichotomy between health and illness, but a concept that describes the process of change in which the state of the mind and body changes continuously between health and illness" (Non-Patent Document 1), and measures to prevent illness are extremely important in order to avoid getting sick.

[0004] However, in the pre-disease state, there are often no ways to know that one is approaching illness (objective findings, subjective symptoms) in the first place. Even if there are ways to know the risk before illness, patients have very little motivation to bear the time and expense of a medical examination to find out the risk when they are not even aware of the illness. Therefore, even if a system is developed that can calculate the risk before contracting an illness, it is extremely difficult to implement the system in society and make it useful for sustainable health maintenance activities.

[0005] Meanwhile, the Ministry of Economy, Trade and Industry has launched the "Health and Productivity Management Excellent Corporation System" since 2015, strengthening incentives for companies that promote "health and productivity management" (Non-Patent Document 3). "Health and productivity management" is a management that monitors employees' health through stress checks, etc., takes measures, maintains the PDCA cycle, and aims to promote organizational health. However, general stress checks have the drawback that many extremely sensitive questions, such as those about relationships with superiors and communication within departments, are difficult for respondents to answer honestly due to consideration for the organization or superiors, and the reliability of the measurement results is low, making it difficult for the measures and PDCA to work effectively. Therefore, for effective "health and productivity management," an evaluation method that can objectively measure unconscious stress is essential.

[0006] On the other hand, fundus examinations and OCT (Optical Coherence Tomography) examinations, which are widely performed in health checkups, are common methods for evaluating the state of retinal blood vessels, and are useful for predicting the risk of retinal diseases such as glaucoma because they are performed regularly in health checkups. It has also been suggested that fundus information resulting from these examinations can contribute to the diagnosis of systemic diseases such as mental disorders, diabetes, and Alzheimer's, in addition to glaucoma, which is a retinal disease (Non-Patent Document 2, Patent Document 1). However, even if it is scientifically possible to predict the risk of these diseases from fundus information, a business model in which patients, the companies to which the patients belong, and the companies that predict risks can all earn profits is essential in order to provide such disease risk prediction services in a sustainable manner for social health maintenance activities.

[0007] On the other hand, according to the Personal Information Protection Law, information obtained from health checkups, including fundus information, is considered sensitive personal information that requires special protection and is subject to stricter regulations. Even if it is a useful service for patients, such as predicting disease risk, if the information is to be used for purposes other than health checkups, consent from all patients must be obtained each time, and this regulation is also a major obstacle to the social implementation of disease risk prediction services. [Prior art documents] [Patent documents]

[0008] [Patent Document 1] No. 6745496 "Diabetic retinopathy stage determination support system and method for supporting diabetic retinopathy stage determination" [Non-patent literature]

[0009] [Non-Patent Document 1] "Preventing illness - a new perspective on health in the era of 100-year lifespans" by Yuji Kuroiwa and Yasuo Otani [Non-Patent Document 2] "Deep learning model using retinal vascular images for classifying schizophrenia" Schizophrenia Research:Vol.241, Page.238-243 (2022) [Non-Patent Document 3] https: / / www.meti.go.jp / policy / mono_info_service / healthcare / downloadfiles / kenkokeiei_gaiyo.pdf Summary of the Invention [Problem to be solved by the invention]

[0010] The objective of the present invention is to provide a system that uses fundus information analysis to calculate a depressive state or risk of depression independently of the subject's consciousness, continuously calculates improvement, worsening, and risk of developing a depressive state or depression, and notifies the calculated values ​​as deviations within the organization (department, company, industry, etc.) to which the subject belongs, thereby supporting health maintenance activities by individuals or organizations and enabling all parties involved to benefit. [Means for solving the problem]

[0011] According to one aspect of the embodiment, the health risk assessment support device includes a prediction processing unit that predicts the health risk of a user based on the user's fundus information, and a communication processing unit that transmits information regarding the health risk of an evaluation subject, including the user. Effect of the Invention

[0012] According to the embodiment, the health risk assessment support device can support health maintenance activities by individuals or organizations by transmitting information on the health risk of an evaluation target. [Brief description of the drawings]

[0013] [Figure 1] FIG. 1 is a block diagram illustrating a configuration of a disease risk assessment support system according to the first embodiment. [Diagram 2] FIG. 2 is a block diagram illustrating the configuration of the server according to the first embodiment. [Diagram 3] FIG. 3 is a diagram illustrating an example of a data structure of data stored in a disease risk storage area of ​​the server according to the first embodiment. [Figure 4] FIG. 4 is a diagram illustrating a flow relating to a deviation of a mental illness risk in the disease risk assessment support system according to the first embodiment. [Diagram 5] FIG. 5 is a flowchart illustrating a process related to a deviation of a mental illness risk in the server according to the first embodiment. [Figure 6] FIG. 6 is a diagram illustrating a flow relating to progression risk in the disease risk assessment support system according to the first embodiment. [Figure 7] FIG. 7 is a flowchart illustrating a process related to a progression risk in the server according to the first embodiment. [Figure 8] FIG. 8 is a diagram illustrating a flow of guidance related to mental illness risk in the disease risk assessment support system according to the first embodiment. [Figure 9]FIG. 9 is a flowchart illustrating a process for guidance related to mental illness risk in the server according to the first embodiment. [Figure 10] FIG. 10 is a diagram illustrating a flow of evaluation of the effect of an applied countermeasure in the disease risk assessment support system according to the first embodiment. [Figure 11] FIG. 11 is a flowchart illustrating a process related to evaluation of the effect of the applied countermeasure in the server according to the first embodiment. [Figure 12] FIG. 12 is a flowchart illustrating a process related to generation of a disease risk prediction model in the server according to the first embodiment. [Figure 13] FIG. 13 is a flowchart illustrating a process related to generation of a progression risk prediction model in the server according to the first embodiment. [Figure 14] FIG. 14 is a block diagram illustrating the configuration of a disease risk assessment support system according to the second embodiment. [Figure 15] FIG. 15 is a block diagram illustrating a configuration of a server according to the second embodiment. [Figure 16] FIG. 16 is a diagram illustrating a flow relating to disease preventive factors in the disease risk assessment support system according to the second embodiment. [Figure 17] FIG. 17 is a flowchart illustrating a process related to disease preventive factors in the server according to the second embodiment. [Figure 18] FIG. 18 is a diagram showing the verification results of the mental disease risk predicted using the disease risk prediction model. [Figure 19] FIG. 19 is a flowchart illustrating a process of adding a function of evaluating a mental illness risk in the server according to the third embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] [First embodiment] Hereinafter, the first embodiment will be described in detail with reference to the drawings. In the drawings, the same components are denoted by the same reference numerals as much as possible, and duplicated explanations will be omitted.

[0015] (Configuration example) FIG. 1 is a block diagram showing an example of the configuration of a disease risk assessment support system 100. The disease risk assessment support system 100 is a system that supports the assessment of mental disease risk of an assessment subject including a user. A mental disease is a disease such as a depressive state or depression. A mental disease is an example of a disease. The disease risk assessment support system 100 is an example of a health risk assessment system. A health risk indicates a risk regarding health.

[0016] The evaluation target including the user is an evaluation unit for evaluating the mental illness risk. The evaluation unit may be an individual or an organization. An organization is a collection to which multiple individuals belong. For example, an organization may be within a range such as an industry, a company, or a department within a company, but is not limited to these. An organization may be larger than an industry, or smaller than a department. The evaluation target including the user not only includes the meaning of the individual user, but also includes the meaning of the organization to which the user belongs. Hereinafter, the evaluation target including the user is also simply referred to as the evaluation target.

[0017] The mental illness risk is a disease risk related to a mental illness. The mental illness risk includes the risk of already suffering from a mental illness at the present time. In this example, the notation "mental illness risk" may be read as "current mental illness risk". The mental illness risk includes the risk of not suffering from a mental illness at the present time, but suffering from a mental illness in the future at the present time. In this example, the notation "mental illness risk" may be read as "future mental illness risk". Risk includes the meaning of possibility. The mental illness risk is data indicating a stage such as a level or score. Hereinafter, the stage is also referred to as an evaluation value. Hereinafter, the mental illness risk will be described as one level selected from a plurality of levels. "Future" may be any time point beyond the present time. The present time refers to the time point at which the mental illness risk is predicted. Hereinafter, the mental illness risk will be described as an example of a disease risk, but the disease risk includes the risk of various diseases such as dementia. The disease risk is an example of a health risk. The health risk includes not only the disease risk but also the state of mental health. The state of mental health is the state of mental health that changes depending on the mental load and stress in daily life. For example, the shock of failing at work is likely to have no effect on the risk of mental illness, but is likely to affect the mental health state. The mental health state is data indicating an evaluation value such as a level or a score. The mental health state may be one level selected from a plurality of levels.

[0018] When the subject of evaluation is an individual user, the mental illness risk of the subject of evaluation is the mental illness risk of the individual user. When the subject of evaluation is an organization, the mental illness risk of the subject of evaluation is the mental illness risk of the organization. The mental illness risk of an organization is a mental illness risk based on the mental illness risks of multiple individuals belonging to the organization. For example, the mental illness risk of an organization may be the average of the mental illness risks of multiple individuals belonging to the organization. For example, the average is an arithmetic mean.

[0019] The disease risk assessment support system 100 includes a server 1, a medical institution terminal 2, and an evaluation subject terminal 3. The server 1, the medical institution terminal 2, and the evaluation subject terminal 3 are communicably connected to each other via a network NW. For example, the network NW includes one or more of various networks such as the Internet, a mobile communication network, and a LAN (Local Area Network). The disease risk assessment support system 100 may also refer to a system including at least two of the server 1, the medical institution terminal 2, and the evaluation subject terminal 3. Although one medical institution terminal 2 is illustrated in FIG. 1, the disease risk assessment support system 100 may include a different medical institution terminal 2 for each medical institution. Although one evaluation subject terminal 3 is illustrated in FIG. 1, the disease risk assessment support system 100 may include a different evaluation subject terminal 3 for each evaluation subject.

[0020] The server 1 is an electronic device that executes a process related to the assessment of mental illness risk. The server 1 is an example of a disease risk assessment support device. The disease risk assessment support device is an example of a health risk assessment support device. An example of the configuration of the server 1 will be described later.

[0021] The medical institution terminal 2 is a terminal that can be used by a medical institution. For example, the medical institution is an institution that performs health checkups on each user and acquires fundus information. The medical institution terminal 2 is an electronic device equipped with a communication function, an information input function, an information display function, and the like. For example, the medical institution terminal 2 is a PC (Personal Computer), a smartphone, a tablet terminal, or the like. The fundus information is information about the fundus. The fundus is a part of the eye that is deeper than the pupil and includes the retina, and the like. The fundus information includes one or both of fundus information of the left eye and fundus information of the right eye. For example, the fundus information is image data related to the fundus. The image data related to the fundus is image data that can be measured by an electronic device. The fundus information may include, as image data related to the fundus, data indicating an image of a cross section of the eye acquired by an OCT device or the like. The fundus information may include, as image data related to the fundus, data indicating a fundus photograph taken by a fundus camera or the like.

[0022] The evaluation target terminal 3 is a terminal that can be used by the evaluation target. The evaluation target terminal 3 is an electronic device equipped with a communication function, an information input function, an information display function, etc. For example, the evaluation target terminal 3 is a PC, a smartphone, a tablet terminal, etc.

[0023] An example of the configuration of the server 1 will be described. FIG. 2 is a block diagram illustrating an example of the configuration of the server 1. The server 1 includes a processor 11, a main memory 12, an auxiliary storage device 13, and a communication interface 14. The components constituting the server 1 are connected to each other so as to be able to input and output signals. In Fig. 2, the interface is written as "I / F".

[0024] The processor 11 corresponds to the central part of the server 1. The processor 11 is an element that constitutes the computer of the server 1. The processor 11 includes one or more processors. For example, the processor is a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit), but is not limited thereto. The processor 11 may be composed of various circuits. The processor loads a program that is pre-stored in the main memory 12 or the auxiliary storage device 13 into the main memory 12. The program is a program for realizing each part described later in the processor 11 of the server 1 and enabling the processor 11 to execute each part. The processor 11 executes the program loaded into the main memory 12 to enable the processing of each part. The processor 11 is an example of a processing circuit.

[0025] The main memory 12 corresponds to the main storage portion of the server 1. The main memory 12 includes a nonvolatile memory area and a volatile memory area. The main memory 12 stores an operating system or a program in the nonvolatile memory area. The main memory 12 uses the volatile memory area as a work area where data is appropriately rewritten by the processor 11. For example, the main memory 12 includes a ROM (Read Only Memory) as a nonvolatile memory area. For example, the main memory 12 includes a RAM (Random Access Memory) as a volatile memory area. The main memory 12 can store a program.

[0026] The auxiliary storage device 13 corresponds to the auxiliary storage portion of the server 1. The auxiliary storage device 13 is an EEPROM (registered trademark) (Electric Erasable Programmable Read-Only Memory), an HDD (Hard Disc Drive), an SSD (Solid State Drive), or the like. The auxiliary storage device 13 can store the above-mentioned programs, data used by the processor 11 in performing various processes, and data generated by the processes of the processor 11. The auxiliary storage device 13 is an example of a storage unit.

[0027] The auxiliary storage device 13 includes a fundus information storage area 130. The fundus information storage area 130 stores fundus information of each of a plurality of individuals. The fundus information storage area 130 may store not only fundus information acquired at one timing for each individual, but also fundus information acquired at each of a plurality of timings. Each of the plurality of timings is a different timing in a chronological order. The plurality of timings may include a timing for each medical checkup once in a predetermined period such as one year. The plurality of timings may include a timing for periodically acquiring fundus information apart from a medical checkup. A group of data including fundus information acquired at each of a plurality of timings is also referred to as time-series data of fundus information. The data stored in the fundus information storage area 130 may be updated every time the server 1 acquires fundus information of any individual.

[0028] The auxiliary storage device 13 includes a disease risk storage area 131. The disease risk storage area 131 stores data indicating the mental illness risk of each of a plurality of evaluation units. The disease risk storage area 131 may store data indicating the mental illness risk of each of a plurality of individuals, or may store data indicating the mental illness risk of each of a plurality of organizations. The disease risk storage area 131 may store not only data indicating the mental illness risk predicted at one timing for each evaluation unit, but also data indicating the mental illness risk predicted at each of a plurality of timings. Each of the multiple timings is a different timing in a time series. A group of data including data indicating the mental illness risk predicted at each of the multiple timings is also referred to as time series data of mental illness risk. Hereinafter, the data indicating the mental illness risk is also referred to as mental illness risk data.

[0029] The disease risk storage area 131 stores data indicating the progression risk of the mental illness risk of each of the multiple evaluation units. The progression risk is the risk of progression of the predicted mental illness risk. For example, the progression risk is the risk of progression from the predicted mental illness risk level to a level that is worse than this level. The progression risk includes the meaning of the possibility of worsening. The progression risk is data indicating an evaluation value such as a level or a score. Here, the progression risk is described as one level selected from multiple levels. When the evaluation target is an individual user, the progression risk of the evaluation target is the progression risk of the individual user. When the evaluation target is an organization, the progression risk of the evaluation target is the progression risk of the organization. The progression risk of the organization is the progression risk based on the progression risks of multiple individuals belonging to the organization. For example, the progression risk of the organization may be the average of the progression risks of multiple individuals belonging to the organization. For example, the average is an arithmetic mean. The disease risk storage area 131 may store data indicating the progression risk of each of the multiple individuals, or may store data indicating the progression risk of each of the multiple organizations. The disease risk storage area 131 may store not only data indicating the progression risk predicted at one timing for each evaluation unit, but also data indicating the progression risk predicted at each of a plurality of timings. Each of the plurality of timings is a different timing in a chronological order. Hereinafter, the data indicating the progression risk is also referred to as progression risk data.

[0030] The data stored in the disease risk storage area 131 may be updated every time the server 1 predicts a mental illness risk of the evaluation subject corresponding to any evaluation unit. The server 1 may predict the mental illness risk of the evaluation subject related to the fundus information at the timing of acquiring the fundus information. The data stored in the disease risk storage area 131 may be updated every time the server 1 predicts a progression risk of the evaluation subject corresponding to any evaluation unit. The server 1 may predict the progression risk at the timing of predicting the mental illness risk. An example of the data structure of the data stored in the disease risk storage area 131 will be described later.

[0031] The auxiliary storage device 13 includes a disease risk prediction model storage area 132. The disease risk prediction model storage area 132 stores a disease risk prediction model. The disease risk prediction model is a model that predicts the user's mental illness risk based on the user's fundus information. The disease risk prediction model is a trained model that is generated based on learning data for disease risk prediction. The disease risk prediction model is an example of a function for evaluating the mental illness risk. The expression "generate" includes not only a newly created mode but also an updated mode. The learning data for disease risk prediction will be described later. The disease risk prediction model outputs the user's mental illness risk data based on the input of the user's fundus information. The disease risk prediction model stored in the disease risk prediction model storage area 132 can be updated every time the server 1 generates a disease risk prediction model.

[0032] The auxiliary storage device 13 includes a learning data storage area 133 for disease risk prediction. The learning data storage area 133 for disease risk prediction stores learning data for disease risk prediction. The learning data for disease risk prediction includes fundus information of each of the multiple learning subjects. The learning data for disease risk prediction includes data on the prevalence of mental illness for each of the multiple learning subjects. The data on the prevalence of mental illness includes data indicating whether or not the learning subjects suffer from a mental illness. The data on the prevalence of mental illness may include data indicating the severity of the mental illness if the learning subjects suffer from a mental illness. The data on the prevalence of mental illness may be data based on a doctor's diagnosis. The data on the prevalence of mental illness may be an index value of mental illness risk. The learning data for disease risk prediction stored in the learning data storage area 133 for disease risk prediction may be updated every time the server 1 acquires data on the learning subjects.

[0033] The auxiliary storage device 13 includes a progression risk prediction model storage area 134. The progression risk prediction model storage area 134 stores a progression risk prediction model. The progression risk prediction model is a model that predicts the progression risk of a user's mental illness risk based on time series data of the user's fundus information. The progression risk prediction model is a trained model generated based on learning data for progression risk prediction. The term "generate" includes not only a new creation mode but also an update mode. The learning data for progression risk prediction will be described later. The progression risk prediction model outputs progression risk data of the user's mental illness risk based on the input of time series data of the user's fundus information. The progression risk prediction model stored in the progression risk prediction model storage area 134 can be updated every time the server 1 generates a progression risk prediction model.

[0034] The auxiliary storage device 13 includes a learning data storage area 135 for predicting a progress risk. The learning data storage area 135 for predicting a progress risk stores learning data for predicting a progress risk. The learning data for predicting a progress risk includes time series data of fundus information of each of the multiple learning subjects. The learning data for predicting a progress risk includes data on the prevalence of a mental illness for each of the multiple learning subjects. The data on the prevalence of a mental illness includes data indicating whether or not the learning subject is suffering from a mental illness. The data on the prevalence of a mental illness may include data indicating the severity of the mental illness if the learning subject is suffering from a mental illness. The data on the prevalence of a mental illness may be data based on a doctor's diagnosis. The learning data for predicting a progress risk stored in the learning data storage area 135 for predicting a progress risk may be updated every time the server 1 acquires data on the learning subject.

[0035] The auxiliary storage device 13 includes a guidance storage area 136. The guidance storage area 136 stores guidance information related to mental illness. The guidance information is information showing guidance of a product or service related to mental illness. The guidance may include the meaning of advertisement or recommendation. The product or service may be related to health promotion that is considered to be effective in preventing mental illness. The guidance information may include information of a uniform resource locator (URL) of the product or service. The guidance storage area 136 stores guidance information according to a mental illness risk. The guidance storage area 136 stores guidance information according to an evaluation unit. For example, the guidance storage area 136 may store different guidance information for each industry. The guidance storage area 136 may store different guidance information for an individual and an organization even in the same industry. The guidance storage area 136 may store different guidance information for a company and a department even in the same industry. The guidance information stored in the guidance storage area 136 may be updated every time the server 1 acquires guidance information.

[0036] The communication interface 14 includes various interfaces that communicably connect the server 1 to other electronic devices via the network NW in accordance with a predetermined communication protocol. The communication interface 14 is an example of a communication unit.

[0037] The hardware configuration of the server 1 is not limited to the above-mentioned configuration. The server 1 allows the above-mentioned components to be omitted or changed and new components to be added as appropriate.

[0038] Each unit realized by the above-mentioned processor 11 will be described. The processor 11 realizes a communication processing unit 110, a prediction processing unit 111, a calculation unit 112, a guidance processing unit 113, an evaluation processing unit 114, and a model processing unit 115. Each unit realized by the processor 11 can also be referred to as each function. Each unit realized by the processor 11 can also be referred to as being realized by a control unit including the processor 11 and the main memory 12.

[0039] The communication processing unit 110 processes the communication between the server 1 and the medical institution terminal 2 via the network NW using the communication interface 14. The communication processing unit 110 processes the communication between the server 1 and the evaluation target terminal 3 via the network NW using the communication interface 14.

[0040] The prediction processing unit 111 predicts the user's mental illness risk based on the user's fundus information. The prediction processing unit 111 predicts the user's mental illness progression risk based on time-series data of the user's fundus information.

[0041] The calculation unit 112 calculates the deviation of the mental illness risk of the evaluation object based on the mental illness risk data of the multiple evaluation units stored in the disease risk memory area 131. The multiple evaluation units include the evaluation object. The multiple evaluation units are a set that includes, in addition to the evaluation object, multiple evaluation units of the same type as the evaluation unit of the evaluation object. The range of the multiple evaluation units can be set appropriately. The deviation is the deviation of the mental illness risk of the evaluation object relative to the set of multiple evaluation units. The deviation is a relative value of the evaluation object relative to a set of standard deviations or percentiles, etc. The deviation is an example of a relative stage of the evaluation object.

[0042] A case where the evaluation target is an individual user will be described. In this example, the multiple evaluation units are multiple individuals. The multiple evaluation units include the individual user of the evaluation target. The multiple evaluation units include individuals who belong to the same organization as the individual user of the evaluation target, in addition to the individual user of the evaluation target. The set of multiple evaluation units is the organization to which the user belongs. The set of multiple evaluation units may be the industry to which the user belongs. In this example, the multiple evaluation units include individuals who belong to the same industry as the individual user of the evaluation target, in addition to the individual user of the evaluation target. The set of multiple evaluation units may be a company to which the user belongs. In this example, the multiple evaluation units include individuals who belong to the same company as the individual user of the evaluation target, in addition to the individual user of the evaluation target. The set of multiple evaluation units may be a department within the company to which the user belongs. In this example, the multiple evaluation units include individuals who belong to the same department as the individual user of the evaluation target, in addition to the individual user of the evaluation target. The deviation of the mental illness risk of the evaluation target is the deviation of the mental illness risk of the individual user. The calculation unit 112 calculates the deviation of the mental illness risk of an individual user with respect to an organization to which the user belongs, which is a collection of multiple individuals, based on the mental illness risk data of multiple individuals stored in the disease risk storage area 131.

[0043] A case where the evaluation target is an organization to which the user belongs will be described. In this example, the multiple evaluation units are multiple organizations. The multiple evaluation units include the organization to which the user belongs. The multiple evaluation units include organizations other than the organization to which the user belongs. The set of multiple evaluation units includes the organization to which the user belongs, and is a range larger than the organization to which the user belongs. When the evaluation target is an industry to which the user belongs, the set of multiple evaluation units includes the industry to which the user belongs, and is a range larger than the industry to which the user belongs. In this example, the multiple evaluation units include industries other than the industry to which the user belongs, which is the evaluation target,. The range larger than the industry to which the user belongs is a range across industries. When the evaluation target is a company to which the user belongs, the set of multiple evaluation units includes the company to which the user belongs, and is a range larger than the company to which the user belongs. In this example, the multiple evaluation units include companies that belong to the same industry as the company to which the user belongs, which is the evaluation target, other than the company to which the user belongs. The range larger than the company to which the user belongs is an industry. The multiple evaluation units may include a company that belongs to a different industry than the company to which the user belongs, in addition to the company to which the user belongs. The range larger than the company to which the user belongs is a range that spans industries. When the evaluation target is a department to which the user belongs, the set of multiple evaluation units includes the department to which the user belongs and is a range larger than the department to which the user belongs. In this example, the multiple evaluation units include other departments that belong to the same company as the department to which the user belongs, in addition to the department to which the evaluation target user belongs. The range larger than the department to which the user belongs is a company. The multiple evaluation units may include departments that belong to a company other than the department to which the evaluation target user belongs, in addition to the department to which the evaluation target user belongs. The company other than the company to which the user belongs may be a company in the same industry as the company to which the user belongs, or may be a company in an industry different from the company to which the user belongs. The department that belongs to a company other than the company to which the user belongs may be a department with the same business as the department to which the user belongs. The range larger than the department to which the user belongs may be an industry, or may be a range that spans industries. The deviation in mental illness risk to be evaluated is the deviation in mental illness risk of the organization to which the user belongs.The calculation unit 112 calculates the deviation of the mental illness risk of the organization to which the user belongs relative to a range larger than the organization to which the user belongs, which is a collection of multiple organizations, based on the mental illness risk data of multiple organizations stored in the disease risk memory area 131.

[0044] The guidance processing unit 113 selects guidance information from the guidance storage area 136 .

[0045] The evaluation processing unit 114 evaluates the effect of the countermeasures applied to the evaluation target on the mental illness risk of the evaluation target based on the time series data of the mental illness risk of the evaluation target. Hereinafter, the countermeasures applied to the evaluation target on the mental illness risk are also referred to as applied countermeasures. The evaluation of the effect of the applied countermeasures is an evaluation of the extent to which the applied countermeasures have an effect on the mental illness risk. The evaluation of the effect of the applied countermeasures includes selecting one stage indicating the effect of the applied countermeasures from multiple stages such as levels or scores. Here, the evaluation of the effect of the applied countermeasures is described as selecting one effect level indicating the effect of the applied countermeasures from multiple effect levels. The multiple effect levels may include one or more effect levels related to "effective", an effect level related to "average effect", and one or more effect levels related to "ineffective". When the mental illness risk is worsening in a time series, the evaluation processing unit 114 can select an effect level related to "ineffective". When the multiple effect levels include multiple effect levels related to "ineffective", the evaluation processing unit 114 can select an effect level related to "ineffective" according to the degree of deterioration of the mental illness risk. The relationship between the degree of worsening of mental illness risk and the multiple effect levels related to "ineffective" can be set as appropriate. When the mental illness risk does not change over time, the evaluation processing unit 114 can select the effect level related to "normal effect". When the mental illness risk improves over time, the evaluation processing unit 114 can select the effect level related to "effective". When the multiple effect levels include multiple effect levels related to "effective", the evaluation processing unit 114 can select the effect level related to "effective" according to the degree of improvement of the mental illness risk. The relationship between the degree of improvement of the mental illness risk and the multiple effect levels related to "effective" can be set as appropriate.

[0046] The model processing unit 115 generates a disease risk prediction model by machine learning. The model processing unit 115 generates a progression risk prediction model by machine learning.

[0047] The data stored in the disease risk storage area 131 will be described. FIG. 3 is a diagram illustrating an example of the data structure of the data stored in the disease risk storage area 131. As shown in FIG. The disease risk storage area 131 stores mental illness risk data and progression risk data for each of a plurality of industries. The disease risk storage area 131 stores, for each company included in each industry, mental illness risk data and progression risk data for each of a plurality of individuals belonging to the company. The disease risk storage area 131 stores mental illness risk data and progression risk data for each company included in each industry. The disease risk storage area 131 stores mental illness risk data and progression risk data for each of a plurality of departments for each company included in each industry.

[0048] (Example of operation) The processing in the disease risk assessment support system 100 will be described. The process steps described below are merely examples, and each process may be modified as much as possible. In addition, steps may be omitted, replaced, or added as appropriate to the process steps described below depending on the embodiment.

[0049] First, the process for deviations in mental illness risk will be described. Here, an example will be described in which an evaluation target includes any user.

[0050] FIG. 4 is a diagram illustrating a flow relating to deviation of mental illness risk in the disease risk assessment support system 100. As shown in FIG.

[0051] The medical institution terminal 2 transmits the fundus information of the user to the server 1 via the network NW. For example, every time the medical institution acquires the fundus information of the user, the medical institution terminal 2 transmits the fundus information of the user to the server 1. The server 1 receives the fundus information of the user from the medical institution terminal 2 via the network NW. The server 1 predicts the mental illness risk of the user based on the fundus information of the user. The server 1 obtains the deviation of the mental illness risk of the evaluation target for a set of multiple evaluation units. The server 1 transmits information indicating the deviation of the mental illness risk of the evaluation target to the medical institution terminal 2 via the network NW. The information indicating the deviation of the mental illness risk of the evaluation target is an example of information regarding the disease risk of the evaluation target. In one example, the server 1 obtains the deviation of the mental illness risk of the user's individual for the organization to which the user belongs, which is a set of multiple individuals. The server 1 transmits information indicating the deviation of the mental illness risk of the user's individual to the medical institution terminal 2 via the network NW. In another example, the server 1 obtains the deviation of the mental illness risk of the organization to which the user belongs for a range larger than the organization to which the user belongs, which is a set of multiple organizations. The server 1 transmits information indicating the deviation of mental illness risk in the organization to which the user belongs to the medical institution terminal 2 via the network NW.

[0052] The medical institution terminal 2 receives information indicating the deviation of the mental illness risk of the evaluation subject from the server 1 via the network NW. The medical institution terminal 2 transmits information indicating the deviation of the mental illness risk of the evaluation subject to the evaluation subject terminal 3 via the network NW. In one example, the medical institution terminal 2 transmits information indicating the deviation of the mental illness risk of the individual user to the personal terminal 3-1 via the network NW. The personal terminal 3-1 refers to a terminal of the evaluation subject terminal 3 that is available to the individual user. In another example, the medical institution terminal 2 transmits information indicating the deviation of the mental illness risk of the organization to which the user belongs to the organization terminal 3-2 via the network NW. The organization terminal 3-2 refers to a terminal of the evaluation subject terminal 3 that is available to the organization.

[0053] FIG. 5 is a flowchart illustrating a process related to deviation of mental illness risk in the server 1.

[0054] The communication processing unit 110 receives the fundus information of the user from the medical institution terminal 2 using the communication interface 14 (step S1). The communication processing unit 110 stores the received fundus information of the user in the fundus information storage area 130 (step S2).

[0055] The prediction processing unit 111 acquires the user's fundus information from the fundus information storage area 130 (step S3). The prediction processing unit 111 acquires the user's fundus information when it is time to predict the mental illness risk. The time to predict the mental illness risk is any timing after the communication processing unit 110 receives the user's fundus information from the medical institution terminal 2.

[0056] The prediction processing unit 111 predicts the mental illness risk of the user based on the fundus information of the user (step S4). In step S4, for example, the prediction processing unit 111 inputs the fundus information of the user to a disease risk prediction model stored in the disease risk prediction model storage area 132. The prediction processing unit 111 acquires the mental illness risk data of the user from the disease risk prediction model based on the input of the fundus information of the user to the disease risk prediction model. The prediction processing unit 111 predicting the mental illness risk of the user is an example of the prediction processing unit 111 predicting the mental illness risk of the evaluation target. The prediction processing unit 111 saves the mental illness risk data of the user in the disease risk storage area 131. The prediction processing unit 111 obtains the mental illness risk of the organization to which the user belongs. For example, the prediction processing unit 111 obtains the mental illness risk of the organization based on the mental illness risks of a plurality of individuals including the user who belongs to the organization. The prediction processing unit 111 obtaining the mental illness risk of the organization is an example of the prediction processing unit 111 predicting the mental illness risk of the evaluation target. The prediction processing unit 111 stores the mental illness risk data of the organization in the illness risk storage area 131.

[0057] The calculation unit 112 calculates the deviation of the mental illness risk of the evaluation target based on the mental illness risk data of the multiple evaluation units stored in the disease risk storage area 131 (step S5). In one example, the calculation unit 112 calculates the deviation of the mental illness risk of the individual user with respect to the organization to which the user belongs, which is a collection of multiple individuals, based on the mental illness risk data of the multiple individuals stored in the disease risk storage area 131. In another example, the calculation unit 112 calculates the deviation of the mental illness risk of the organization to which the user belongs, which is a collection of multiple organizations, with respect to a range larger than the organization to which the user belongs, which is a collection of multiple organizations, based on the mental illness risk data of the multiple organizations stored in the disease risk storage area 131.

[0058] The communication processing unit 110 transmits information indicating the deviation of the mental illness risk of the evaluation target to the medical institution terminal 2 using the communication interface 14 (step S6). In one example, the communication processing unit 110 transmits information indicating the deviation of the mental illness risk of the individual user to the medical institution terminal 2. The medical institution terminal 2 transmits information indicating the deviation of the mental illness risk of the individual user to the personal terminal 3-1. In another example, the communication processing unit 110 transmits information indicating the deviation of the mental illness risk of the organization to which the user belongs to the medical institution terminal 2. The medical institution terminal 2 transmits information indicating the deviation of the mental illness risk of the organization to which the user belongs to the organization terminal 3-2. The communication processing unit 110 may transmit information indicating the deviation of the mental illness risk of the evaluation target to the evaluation target terminal 3 using the communication interface 14. In one example, the communication processing unit 110 may transmit information indicating the deviation of the mental illness risk of the individual user to the personal terminal 3-1. In another example, the communication processing unit 110 may transmit information indicating the deviation of the mental illness risk of the organization to which the user belongs to the organization terminal 3-2.

[0059] The communication processing unit 110 may transmit information indicating the mental illness risk of the evaluation subject to the medical institution terminal 2 using the communication interface 14. The information indicating the mental illness risk of the evaluation subject is an example of information regarding the disease risk of the evaluation subject. In one example, the communication processing unit 110 transmits information indicating the mental illness risk of the user's personal to the medical institution terminal 2. In another example, the communication processing unit 110 transmits information indicating the mental illness risk of the organization to which the user belongs to the medical institution terminal 2. The communication processing unit 110 may transmit information indicating the mental illness risk of the evaluation subject to the evaluation subject terminal 3 using the communication interface 14. In one example, the communication processing unit 110 may transmit information indicating the mental illness risk of the user's personal to the personal terminal 3-1. In another example, the communication processing unit 110 may transmit information indicating the mental illness risk of the organization to which the user belongs to the organization terminal 3-2.

[0060] Next, the process related to the progressing risk will be described. Here, an example of an evaluation target including any user will be described.

[0061] FIG. 6 is a diagram illustrating a flow relating to progression risk in the disease risk assessment support system 100. As shown in FIG.

[0062] The server 1 predicts the user's risk of progression of mental illness based on time-series data of the user's fundus information. The server 1 transmits information on the progression risk of the evaluation subject to the medical institution terminal 2 via the network NW. For example, the information on the progression risk of the evaluation subject is information indicating the progression risk of the evaluation subject. The information on the progression risk of the evaluation subject is an example of information on the mental illness risk of the evaluation subject. In one example, the server 1 transmits information on the user's personal progression risk to the medical institution terminal 2 via the network NW. In another example, the server 1 transmits information on the progression risk of the organization to which the user belongs to the medical institution terminal 2 via the network NW.

[0063] The medical institution terminal 2 receives information on the progression risk of the evaluation subject from the server 1 via the network NW. The medical institution terminal 2 transmits the information on the progression risk of the evaluation subject to the evaluation subject terminal 3 via the network NW. In one example, the medical institution terminal 2 transmits information on the user's personal progression risk to the personal terminal 3-1 via the network NW. In another example, the medical institution terminal 2 transmits information on the progression risk of the organization to which the user belongs to the organization terminal 3-2 via the network NW.

[0064] FIG. 7 is a flowchart illustrating a process related to a progression risk in the server 1. As shown in FIG.

[0065] The prediction processing unit 111 acquires time series data of the user's fundus information from the fundus information storage area 130 (step S11). When it is time to predict the progression risk, the prediction processing unit 111 acquires the time series data of the user's fundus information. The time to predict the progression risk is the time to predict the mental illness risk. The time series data of the user's fundus information includes the user's fundus information used to predict the user's mental illness risk.

[0066] The prediction processing unit 111 predicts the progression risk of the user's disease risk based on the time series data of the user's fundus information (step S12). In step S12, for example, the prediction processing unit 111 inputs the time series data of the user's fundus information to the progression risk prediction model stored in the progression risk prediction model storage area 134. The prediction processing unit 111 acquires the user's progression risk data from the progression risk prediction model based on the input of the time series data of the user's fundus information to the progression risk prediction model. The prediction processing unit 111 predicting the user's progression risk is an example of the prediction processing unit 111 predicting the progression risk of the evaluation target. The prediction processing unit 111 saves the user's progression risk data in the disease risk storage area 131. The prediction processing unit 111 obtains the progression risk of the organization to which the user belongs. For example, the prediction processing unit 111 obtains the progression risk of the organization based on the progression risks of multiple individuals including the user who belongs to the organization. The prediction processing unit 111 obtaining the progression risk of the organization is an example of the prediction processing unit 111 predicting the progression risk of the evaluation target. The prediction processing unit 111 stores the progression risk of the tissue in the disease risk storage area 131.

[0067] The communication processing unit 110 transmits information on the progression risk of the evaluation subject to the medical institution terminal 2 using the communication interface 14 (step S13). In one example, the communication processing unit 110 transmits information on the user's personal progression risk to the medical institution terminal 2. The medical institution terminal 2 transmits information on the user's personal progression risk to the personal terminal 3-1. In another example, the communication processing unit 110 transmits information on the progression risk of the organization to which the user belongs to the medical institution terminal 2. The medical institution terminal 2 transmits information on the progression risk of the organization to which the user belongs to the organization terminal 3-2. The communication processing unit 110 may transmit information on the progression risk of the evaluation subject to the evaluation subject terminal 3 using the communication interface 14. In one example, the communication processing unit 110 may transmit information on the user's personal progression risk to the personal terminal 3-1. In another example, the communication processing unit 110 may transmit information on the progression risk of the organization to which the user belongs to the organization terminal 3-2.

[0068] Next, the process of providing guidance related to mental illness risk will be described. Here, an example will be described in which an evaluation target includes any user.

[0069] FIG. 8 is a diagram illustrating an example of a flow of guidance related to mental illness risk in the disease risk assessment support system 100. As shown in FIG.

[0070] The server 1 predicts the user's mental illness risk based on the user's fundus information. The server 1 transmits guidance information related to the mental illness risk of the evaluation target to the medical institution terminal 2 via the network NW. In one example, the server 1 transmits guidance information related to the user's personal mental illness risk to the medical institution terminal 2 via the network NW. In another example, the server 1 transmits guidance information related to the mental illness risk of the organization to which the user belongs to the medical institution terminal 2 via the network NW.

[0071] The medical institution terminal 2 receives guidance information related to the mental illness risk of the evaluation subject from the server 1 via the network NW. The medical institution terminal 2 transmits guidance information related to the mental illness risk of the evaluation subjects including the user to the evaluation subject terminal 3 via the network NW. In one example, the medical institution terminal 2 transmits guidance information related to the user's personal mental illness risk to the personal terminal 3-1 via the network NW. In another example, the medical institution terminal 2 transmits guidance information related to the mental illness risk of the organization to which the user belongs to the organization terminal 3-2 via the network NW.

[0072] FIG. 9 is a flowchart illustrating a process for providing guidance related to mental illness risk in the server 1.

[0073] The prediction processing unit 111 acquires the fundus information of the user from the fundus information storage area 130 (step S21). The process of step S21 is similar to the process of step S3 described above. The prediction processing unit 111 predicts the user's mental illness risk based on the user's fundus information (step S22). The process of step S22 is similar to the process of step S4 described above. Based on the mental illness risk of the evaluation target, the guidance processing unit 113 selects guidance information related to the mental illness risk of the evaluation target from the guidance memory area 136 (step S23). In one example, based on the mental illness risk of the user's personal, the guidance processing unit 113 selects guidance information related to the mental illness risk of the user's personal from the guidance memory area 136. In another example, based on the mental illness risk of the user's personal organization, the guidance processing unit 113 selects guidance information related to the mental illness risk of the user's organization from the guidance memory area 136.

[0074] The guidance information selected by the guidance processing unit 113 may differ depending on the level of mental illness risk of the evaluation target. This is because the countermeasures vary depending on the level of mental illness risk, and therefore the necessary products or services also vary depending on the level of mental illness risk. The guidance information selected by the guidance processing unit 113 may differ depending on the type of evaluation unit, even if the level of mental illness risk is the same. For example, the guidance information differs depending on whether the guidance information is notified to the user or the organization to which the user belongs. This is because the possible countermeasures vary between an individual and an organization, and therefore the necessary products or services also vary depending on the evaluation unit.

[0075] The communication processing unit 110 transmits guidance information related to the mental illness risk of the evaluation subjects, including the user, to the medical institution terminal 2 using the communication interface 14 (step S24). In one example, the communication processing unit 110 transmits guidance information related to the user's personal mental illness risk to the medical institution terminal 2. The medical institution terminal 2 transmits guidance information related to the user's personal mental illness risk to the personal terminal 3-1. In another example, the communication processing unit 110 transmits guidance information related to the mental illness risk of the organization to which the user belongs to the medical institution terminal 2. The medical institution terminal 2 transmits guidance information related to the mental illness risk of the organization to which the user belongs to the organization terminal 3-2. The communication processing unit 110 may transmit guidance information related to the mental illness risk of the evaluation subjects, including the user, to the evaluation subject terminal 3 using the communication interface 14. In one example, the communication processing unit 110 may transmit guidance information related to the user's personal mental illness risk to the personal terminal 3-1. In another example, the communication processing unit 110 may transmit guidance information related to the mental illness risk of the organization to which the user belongs to the organization terminal 3-2.

[0076] The guidance information may be insurance information related to insurance. The insurance is insurance that insures the evaluation target against a state caused by a mental illness risk. For example, the insurance is insurance that insures the evaluation target against an inability to work, but is not limited to this. In this example, in step S23, the guidance processing unit 113 selects insurance information from the guidance memory area 136 based on the mental illness risk of the evaluation target. In one example, the guidance processing unit 113 selects insurance information related to the mental illness risk of the individual user from the guidance memory area 136 based on the mental illness risk of the individual user. In another example, the guidance processing unit 113 selects insurance information related to the mental illness risk of the organization to which the user belongs from the guidance memory area 136 based on the mental illness risk of the organization to which the user belongs.

[0077] The insurance information selected by the guidance processing unit 113 may differ depending on the level of mental illness risk of the subject to be evaluated. This is because the need for insurance changes depending on the level of mental illness risk. At least one of the insurance content and insurance premium indicated by the insurance information may differ depending on the level of mental illness risk of the subject to be evaluated. The insurance content may become more generous as the level of mental illness risk of the subject to be evaluated increases. This is because the period until the mental illness risk improves and the possibility of employment interference change depending on the level of mental illness risk. The insurance premium may become cheaper as the level of mental illness risk of the subject to be evaluated decreases. This is because the possibility of the mental illness worsening or the possibility of contracting a mental illness changes depending on the level of mental illness risk.

[0078] In step S24, the communication processing unit 110 transmits insurance information related to the mental illness risk to the medical institution terminal 2 using the communication interface 14 based on the mental illness risk of the evaluation target. In one example, the communication processing unit 110 transmits insurance information related to the user's personal mental illness risk to the medical institution terminal 2. The medical institution terminal 2 transmits insurance information related to the user's personal mental illness risk to the personal terminal 3-1. In another example, the communication processing unit 110 transmits insurance information related to the mental illness risk of the organization to which the user belongs to the medical institution terminal 2. The medical institution terminal 2 transmits insurance information related to the mental illness risk of the organization to which the user belongs to the organization terminal 3-2. The communication processing unit 110 may transmit the insurance information to the evaluation target terminal 3 using the communication interface 14. In one example, the communication processing unit 110 may transmit insurance information related to the user's personal mental illness risk to the personal terminal 3-1. In another example, the communication processing unit 110 may transmit insurance information related to the mental illness risk of the organization to which the user belongs to the organization terminal 3-2.

[0079] Next, the process related to the applied measures will be described. Here, an example will be described in which an evaluation target including an arbitrary user is used. The evaluation target may be an individual user or an organization to which the user belongs.

[0080] FIG. 10 is a diagram illustrating a flow of evaluation of the effect of an applied countermeasure in the disease risk assessment support system 100. As shown in FIG.

[0081] The evaluation subject terminal 3 transmits the applied countermeasure information to the server 1 via the network NW. The applied countermeasure information is information related to the applied countermeasure. The applied countermeasure information may include information that can identify the content of the applied countermeasure. The applied countermeasure information may include information indicating the period of the applied countermeasure. The server 1 receives the applied countermeasure information from the evaluation subject terminal 3 via the network NW. The server 1 evaluates the effectiveness of the applied countermeasure based on the time series data of the mental illness risk of the evaluation subject. The server 1 transmits evaluation information to the evaluation subject terminal 3 via the network NW. The evaluation information is information indicating an evaluation of the effectiveness of the applied countermeasure. The evaluation information may be information indicating one effect level selected from a plurality of effect levels. The evaluation subject terminal 3 receives the evaluation information from the server 1 via the network NW. The evaluation subject terminal 3 can receive the evaluation information from the server 1 every time it transmits the applied countermeasure information to the server 1. The evaluation subject can understand which countermeasure is appropriate for him / herself by comparing the evaluation information.

[0082] FIG. 11 is a flowchart illustrating a process related to evaluation of the effectiveness of the applied countermeasure in the server 1.

[0083] The communication processing unit 110 transmits countermeasure information regarding countermeasures for the mental illness risk of the evaluation subject to the medical institution terminal 2 based on the mental illness risk of the evaluation subject using the communication interface 14 (step S31). The countermeasure information is information indicating countermeasures suitable for avoiding and reducing the mental illness risk. The countermeasure information may include information indicating products suitable for avoiding and reducing the mental illness risk as countermeasures. The countermeasure information may include information indicating lifestyle habits suitable for avoiding and reducing the mental illness risk as countermeasures. The countermeasure information may differ according to the level of the mental illness risk of the evaluation subject. This is because the appropriate countermeasure changes according to the level of the mental illness risk. For example, the strength of the countermeasure differs when the level of the mental illness risk is equal to or higher than a predetermined level and when it is lower than a predetermined level. The auxiliary storage device 13 stores countermeasure information according to the level of the mental illness risk. Even if the level of the mental illness risk is the same, it may differ according to the type of evaluation unit. This is because the appropriate countermeasure changes between an individual and an organization. When the evaluation unit is an individual, the countermeasure may be one that can be taken by the individual. When the evaluation unit is an organization, the countermeasure may be one that can be taken by the organization. The auxiliary storage device 13 stores the countermeasure information according to the type of evaluation unit. The medical institution terminal 2 transmits the countermeasure information to the evaluation target terminal 3 via the network NW. The communication processing unit 110 may transmit the countermeasure information to the evaluation target terminal 3 using the communication interface 14.

[0084] The communication processing unit 110 receives the applied measure information from the evaluation target terminal 3 using the communication interface 14 (step S32). Note that the communication processing unit 110 may also receive the applied measure information from the medical institution terminal 2 using the communication interface 14.

[0085] The evaluation processing unit 114 acquires time-series data on the mental illness risk of the evaluation target from the illness risk storage area 131 (step S33). The time-series data on the mental illness risk is data within the period of the applied countermeasure.

[0086] The evaluation processing unit 114 evaluates the effectiveness of the applied countermeasure based on the time-series data of the mental illness risk of the evaluation target (step S34).

[0087] The communication processing unit 110 transmits the evaluation information to the evaluation target terminal 3 using the communication interface 14 (step S35). The communication processing unit 110 may transmit the evaluation information to the medical institution terminal 2 using the communication interface 14. The medical institution terminal 2 transmits the evaluation information to the evaluation target terminal 3 via the network NW.

[0088] Next, the process for generating a disease risk prediction model will be described.

[0089] FIG. 12 is a flowchart illustrating a process for generating a disease risk prediction model in the server 1. The model processing unit 115 may start generating a disease risk prediction model at any timing and create a new disease risk prediction model. The model processing unit 115 may start generating a disease risk prediction model at any timing and update the disease risk prediction model.

[0090] The model processing unit 115 acquires disease risk prediction learning data from the disease risk prediction learning data storage area 133 (step S41).

[0091] The model processing unit 115 generates a disease risk prediction model by machine learning based on the learning data for disease risk prediction (step S42). In step S42, for example, the model processing unit 115 estimates the relationship of the fundus information to the data on the morbidity of a mental illness by machine learning. The model processing unit 115 generates a disease risk prediction model based on the estimation. The machine learning is, but is not limited to, a neural network or the like. When a person suffers from a mental illness, the fundus information tends to have different characteristics compared to when the person does not suffer from a mental illness. The fundus information tends to have different characteristics depending on the degree of the mental illness. There may be a certain correlation between the fundus information and the data on the morbidity of a mental illness. Therefore, there may also be a certain correlation between the fundus information and the risk of a mental illness. In one example, there may be a certain correlation between the fundus information and the current risk of a mental illness. In another example, there may be a certain correlation between the fundus information and the risk of a mental illness in the future.

[0092] The model processing unit 115 stores the generated disease risk prediction model in the disease risk prediction model storage area 132 (step S43).

[0093] FIG. 18 is a diagram showing the verification results of the mental disease risk predicted using the disease risk prediction model. Here, the Patient Health Questionnaire-9 (PHQ-9), which is used to diagnose depression, is used as an index of mental illness risk. The horizontal axis indicates the actual PHQ-9 score of the subject. The vertical axis indicates the PHQ-9 score of the subject predicted using the disease risk prediction model. The TP field indicates the area where the actual PHQ-9 score is 5 or more and the predicted PHQ-9 score is also 5 or more. The TN field indicates the area where the actual PHQ-9 score is less than 5 and the predicted PHQ-9 score is also less than 5. The FP field indicates the area where the actual PHQ-9 score is less than 5 but the predicted PHQ-9 score is 5 or more. The FN field indicates the area where the actual PHQ-9 score is 5 or more but the predicted PHQ-9 score is less than 5. The double circle marks indicate the data of each of the 242 subjects. It can be seen that the predicted PHQ-9 score tends to match or approximately match the actual PHQ-9 score. Therefore, it can be said that there may be a certain correlation between the fundus information contained in the disease risk prediction learning data used to generate a disease risk prediction model and data regarding the incidence of mental illness.

[0094] Next, the process for generating a progression risk prediction model will be described.

[0095] FIG. 13 is a flowchart illustrating a process related to generation of a progression risk prediction model in the server 1. The model processing unit 115 may start generating a progression risk prediction model at any timing and create a new progression risk prediction model. The model processing unit 115 may start generating a progression risk prediction model at any timing and update the progression risk prediction model.

[0096] The model processing unit 115 acquires the progression risk prediction learning data from the progression risk prediction learning data storage area 135 (step S51).

[0097] The model processing unit 115 generates a progression risk prediction model by machine learning based on the progression risk prediction learning data (step S52). In step S52, for example, the model processing unit 115 estimates the relationship of the time series data of fundus information to the data on the affliction of a mental illness by machine learning. The model processing unit 115 generates a progression risk prediction model based on the estimation. The machine learning is, but is not limited to, a neural network or the like. When a person suffers from a mental illness, the fundus information tends to show different characteristics compared to when the person does not suffer from a mental illness. The fundus information tends to show different characteristics depending on the degree of the mental illness. The characteristics that appear in the fundus information tend to change in time series. There may be a certain correlation between the time series data of the fundus information and the data on the affliction of a mental illness. Therefore, there may also be a certain correlation between the time series data of the fundus information and the progression risk.

[0098] The model processing unit 115 stores the generated progress risk prediction model in the progress risk prediction model storage area 134 (step S53).

[0099] (Modification) The functions of each of the above-mentioned parts realized by the processor 11 of the server 1 may be distributed between the server 1 and the medical institution terminal 2. In this example, the medical institution terminal 2 includes a processor, a main memory, an auxiliary storage device, and a communication interface. The processor, main memory, auxiliary storage device, and communication interface of the medical institution terminal 2 are configured in the same manner as the processor 11, main memory 12, auxiliary storage device 13, and communication interface 14 of the server 1. The processor is an example of a processing circuit. The auxiliary storage device is an example of a memory unit. The communication interface is an example of a communication unit.

[0100] The processor of the medical institution terminal 2 realizes a part of the communication processing unit, the prediction processing unit, the calculation unit, the guidance processing unit, the evaluation processing unit, and the model processing unit. The processing of the communication processing unit, the prediction processing unit, the calculation unit, the guidance processing unit, the evaluation processing unit, and the model processing unit realized by the processor of the medical institution terminal 2 may be similar to the processing of the communication processing unit 110, the prediction processing unit 111, the calculation unit 112, the guidance processing unit 113, the evaluation processing unit 114, and the model processing unit 115 realized by the processor 11 of the server 1, and can be interpreted as being interchangeable.

[0101] The auxiliary storage device of the medical institution terminal 2 may store fundus information of each of a plurality of individuals, similar to the auxiliary storage device 13 of the server 1. The auxiliary storage device of the medical institution terminal 2 may store mental illness risk data of each of a plurality of evaluation units, similar to the auxiliary storage device 13 of the server 1. The auxiliary storage device of the medical institution terminal 2 may store progression risk data of mental illness risk of each of a plurality of evaluation units, similar to the auxiliary storage device 13 of the server 1. The auxiliary storage device of the medical institution terminal 2 may store a disease risk prediction model, similar to the auxiliary storage device 13 of the server 1. The auxiliary storage device of the medical institution terminal 2 may store learning data for disease risk prediction, similar to the auxiliary storage device 13 of the server 1. The auxiliary storage device of the medical institution terminal 2 may store a progression risk prediction model, similar to the auxiliary storage device 13 of the server 1. The auxiliary storage device of the medical institution terminal 2 may store learning data for progression risk prediction, similar to the auxiliary storage device 13 of the server 1. The auxiliary storage device of the medical institution terminal 2 may store guidance information related to mental illness, similar to the auxiliary storage device 13 of the server 1.

[0102] For example, the processor of the medical institution terminal 2 realizes at least a communication processing unit and a prediction processing unit, but may not realize a calculation unit. The processor 11 of the server 1 may realize a calculation unit 112. In this example, the communication processing unit of the medical institution terminal 2 transmits mental illness risk data indicating the mental illness risk of the evaluation subject predicted by the prediction processing unit to the server 1 using the communication interface. The communication processing unit of the medical institution terminal 2 receives information indicating the deviation of the mental illness risk of the evaluation subject obtained by the calculation unit 112 of the server 1 from the server 1 using the communication interface. The communication processing unit of the medical institution terminal 2 transmits information indicating the deviation of the mental illness risk of the evaluation subject to the evaluation subject terminal 3 using the communication interface. The communication processing unit of the medical institution terminal 2 transmits information indicating the mental illness risk of the evaluation subject predicted by the prediction processing unit to the evaluation subject terminal 3 using the communication interface. The communication processing unit of the medical institution terminal 2 transmits information regarding the progression risk of the evaluation subject predicted by the prediction processing unit to the evaluation subject terminal 3 using the communication interface. As described above, the medical institution terminal 2 provides the mental illness risk of the evaluation subject to an external institution different from the medical institution. When the evaluation target is an organization, the mental illness risk of the evaluation target is statistical information such as the average mental illness risk of multiple individuals belonging to the organization. Statistical information is not subject to the regulations of the Personal Information Protection Act. Therefore, in this example, it is possible to operate with consideration given to the handling of personal information.

[0103] For example, the processor of the medical institution terminal 2 realizes at least a communication processing unit and a prediction processing unit, but does not necessarily have to realize a guidance processing unit. The processor 11 of the server 1 may realize a guidance processing unit 113. In this example, the communication processing unit of the medical institution terminal 2 uses the communication interface to transmit mental illness risk data indicating the mental illness risk of the evaluation subject predicted by the prediction processing unit to the server 1. The communication processing unit of the medical institution terminal 2 uses the communication interface to receive from the server 1 guidance information related to the mental illness risk of the evaluation subject selected by the guidance processing unit 113 of the server 1. The communication processing unit of the medical institution terminal 2 transmits guidance information related to the mental illness risk of the evaluation subject to the evaluation subject terminal 3 using the communication interface.

[0104] For example, the processor of the medical institution terminal 2 realizes at least a communication processing unit, but may not realize a prediction processing unit, a calculation unit, and a guidance processing unit. The processor 11 of the server 1 may realize a prediction processing unit 111, a calculation unit 112, and a guidance processing unit 113. In this example, the communication processing unit of the medical institution terminal 2 transmits the fundus information of the user to the server 1 using the communication interface. The communication processing unit of the medical institution terminal 2 receives information indicating the deviation of the mental illness risk of the evaluation subject calculated by the calculation unit 112 of the server 1 from the server 1 using the communication interface. The communication processing unit of the medical institution terminal 2 transmits information indicating the deviation of the mental illness risk of the evaluation subject to the evaluation subject terminal 3 using the communication interface. The communication processing unit of the medical institution terminal 2 receives information indicating the mental illness risk of the evaluation subject predicted by the prediction processing unit 111 of the server 1 from the server 1 using the communication interface. The communication processing unit of the medical institution terminal 2 transmits information indicating the mental illness risk of the evaluation subject to the evaluation subject terminal 3 using the communication interface. The communication processing unit of the medical institution terminal 2 receives, using the communication interface, information on the progression risk of the evaluation subject predicted by the prediction processing unit 111 of the server 1 from the server 1. The communication processing unit of the medical institution terminal 2 transmits, using the communication interface, information on the progression risk of the evaluation subject to the evaluation subject terminal 3. The communication processing unit of the medical institution terminal 2 receives, using the communication interface, guidance information related to the mental illness risk of the evaluation subject selected by the guidance processing unit 113 of the server 1 from the server 1. The communication processing unit of the medical institution terminal 2 transmits, using the communication interface, guidance information related to the mental illness risk of the evaluation subject to the evaluation subject terminal 3.

[0105] Although several examples have been described in which the functions of the above-mentioned units implemented by the processor 11 of the server 1 are distributed between the server 1 and the medical institution terminal 2, the manner of distribution is not limited to these.

[0106] The server 1 may be an electronic device that can be used by a medical institution, similar to the medical institution terminal 2.

[0107] (effect) The health risk assessment support device includes a prediction processing unit that predicts a health risk of a user based on fundus information of the user. The health risk assessment support device includes a communication processing unit that transmits information regarding the health risk of an evaluation target including the user. The health risk assessment support device can obtain the health risk of the subject to be evaluated, and can support health maintenance activities by individuals or organizations by transmitting information regarding the health risk of the subject to be evaluated. The subject to be evaluated can grasp health risks before their health deteriorates, such as before the disease to which the individual or each person belonging to the organization suffers worsens, before they contract a disease, or before their mental health condition deteriorates. The subject to be evaluated can take appropriate measures before the health of the individual or each person belonging to the organization deteriorates. The subject to be evaluated can also grasp the progress of health risks.

[0108] The prediction processing unit predicts a progression risk of a health risk of the user based on time-series data of the fundus information of the user. The communication processing unit transmits information regarding the progression risk of the evaluation target. The health risk assessment support device can obtain the progression risk of the subject to be evaluated, and can support health maintenance activities by individuals or organizations by transmitting information regarding the progression risk of the subject to be evaluated. The subject to be evaluated can grasp the progression risk before the health of the individual or each person belonging to the organization deteriorates. The subject to be evaluated can take appropriate measures before the health of the individual or each person belonging to the organization deteriorates. The subject to be evaluated can also grasp the progress of the progression risk.

[0109] The health risk assessment support device includes a calculation unit that calculates a deviation of a health risk of an evaluation target based on health risk data of a plurality of evaluation units stored in a storage unit. The information on the health risk of the evaluation target includes information indicating the deviation of the health risk of the evaluation target. The evaluation target is an organization to which the user belongs. The set of multiple evaluation units is an organization that includes the organization to which the user belongs and is larger than the organization to which the user belongs. The deviation of the health risk of the evaluation target is the deviation of the health risk of the evaluation target with respect to the set of multiple evaluation units. The evaluation target is an individual user. The set of multiple evaluation units is an organization to which the user belongs. The deviation of the health risk of the evaluation target is the deviation of the health risk of the evaluation target with respect to the set of multiple evaluation units. The health risk assessment support device can obtain the deviation of the health risk of the subject to be evaluated, and can support health maintenance activities by individuals or organizations by providing information indicating the deviation of the health risk of the subject to be evaluated. The subject to be evaluated can grasp the relative level of the health risk by grasping the deviation of the health risk of the subject to be evaluated before the health of the individual or each person belonging to the organization deteriorates. The subject to be evaluated can take appropriate measures before the health of the subject deteriorates according to the relative level of the health risk. The subject to be evaluated can also grasp the transition of the deviation of the health risk.

[0110] The communication processing unit transmits countermeasure information regarding countermeasures for the health risk of the evaluation target, based on the health risk of the evaluation target. The health risk assessment support device can support health maintenance activities for avoiding and reducing health risks by providing countermeasure information based on health risks.

[0111] Countermeasure information varies depending on the level of health risk being assessed. The health risk assessment support device can provide countermeasure information indicating countermeasures according to the level of health risk.

[0112] The health risk assessment support device includes an evaluation processing unit that evaluates the effectiveness of a measure applied to the evaluation target against the health risk of the evaluation target based on time-series data of the health risk of the evaluation target. The health risk assessment support device can support the subject of evaluation in taking appropriate measures by evaluating the effects of applied measures. The subject of evaluation can grasp the effects of applied measures. The subject of evaluation can grasp which measures are appropriate for the subject of evaluation by comparing the evaluation information for each applied measure.

[0113] The communication processing unit transmits guidance information related to the health risk of the evaluation target. The health risk assessment support device can support health maintenance activities by individuals or organizations by recommending products or services that are effective in preventing disease or products or services that are effective for mental health using guidance information related to the health risk of the subject of evaluation.

[0114] The guidance information varies depending on the level of health risk of the subject to be evaluated. The health risk assessment support device can provide guidance information that recommends necessary products or services depending on the level of health risk.

[0115] The guidance information differs depending on whether the guidance information is to be notified to the user or to the organization to which the user belongs. The health risk assessment assistance device can provide guidance information that recommends products or services necessary for the destination of the guidance information.

[0116] The communication processing unit transmits insurance information regarding insurance that provides insuring incapacity to work to the subject to be evaluated based on the subject's health risk. The health risk assessment support device can support health maintenance activities by individuals or organizations by recommending insurance that covers inability to work.

[0117] At least one of the insurance content and the premium varies depending on the level of health risk of the subject to evaluation. The health risk assessment support device can provide necessary insurance information according to the level of health risk.

[0118] [Second embodiment] The second embodiment will be described in detail below with reference to the drawings. The same components as those in the first embodiment are denoted by the same reference numerals as much as possible, and duplicated descriptions will be omitted.

[0119] (Configuration example) 14 is a block diagram illustrating an example of the configuration of a disease risk assessment support system 200. The disease risk assessment support system 200 is a system that supports the assessment of a mental disease risk of an assessment subject.

[0120] The disease risk assessment support system 200 includes a server 1, a medical institution terminal 2, and a corporate terminal 4. The disease risk assessment support system 200 is an example of a health risk assessment system. The server 1, the medical institution terminal 2, and the corporate terminal 4 are communicably connected to each other via a network NW. The disease risk assessment support system 100 may also refer to a system including at least two of the server 1, the medical institution terminal 2, and the corporate terminal 4. Although one medical institution terminal 2 is illustrated in FIG. 14, the disease risk assessment support system 200 may include a different medical institution terminal 2 for each medical institution. The server 1 is an example of a disease risk assessment support device. The disease risk assessment support device is an example of a health risk assessment support device.

[0121] The corporate terminal 4 is a terminal that can be used by a company. The corporate terminal 4 is an electronic device that has a communication function, an information input function, an information display function, etc. For example, the corporate terminal 4 is a PC, a smartphone, a tablet terminal, etc.

[0122] An example of the configuration of the server 1 will be described. FIG. 15 is a block diagram illustrating an example of the configuration of the server 1. The server 1 includes a processor 11, a main memory 12, an auxiliary storage device 13, and a communication interface 14. The components constituting the server 1 are connected to each other so as to be able to input and output signals. In Fig. 15, the interface is written as "I / F".

[0123] The auxiliary storage device 13 includes a fundus information storage area 130, a disease risk storage area 131, a disease risk prediction model storage area 132, and a learning data storage area 133 for disease risk prediction, similarly to the first embodiment.

[0124] The auxiliary storage device 13 includes an attribute storage area 137. The attribute storage area 137 stores data indicating the attributes of each of a plurality of individuals. Hereinafter, the data indicating the attributes is also referred to as attribute data. The attribute data is data indicating the characteristics of an individual. For example, the attribute data includes at least one of data regarding lifestyle habits, biological data, and environmental data. The data regarding lifestyle habits may be data that can be acquired by a questionnaire for each person. The questionnaire may be conducted at the timing of each health checkup or at any time. The data regarding lifestyle habits includes data on various elements. The data regarding lifestyle habits may include data such as sleep time, work hours, alcohol consumption, smoking amount, exercise time, and dietary habits per day. The biological data may be data that can be measured from a living body by measurement. The measurement may be conducted at the timing of each health checkup or at any time. The biological data includes data on various elements. The biological data may include data based on values ​​that can be measured from a living body, such as height, weight, and biomarkers such as blood pressure, blood analysis results, and biological tissue analysis results. The data based on the value may be data indicating a measured value or data obtained by converting a measured value. In one example, the data obtained by converting a measured value is data indicating a value of another index obtained by converting the values ​​of two or more measured indexes. The data obtained by converting a measured value may be data indicating a value of a Body Mass Index (BMI) converted based on a height value and a weight value. In another example, the data obtained by converting a measured value is data indicating any one of a plurality of hierarchies corresponding to the value of the measured index. The data obtained by converting a measured value may be data indicating any one of three hierarchies, high, medium, or low, corresponding to the value of the measured biomarker. The biological data may include image data that can be measured by an electronic device. The image data may include data indicating a photograph taken by a camera. The image data may include image data of the inside of a living body obtained by CT (Computed Tomography) and MRI (Magnetic Resonance Imaging), etc. The image data of the inside of a living body is data indicating an image of a cross section of the living body. The environmental data is data related to the living environment in which each person is placed.The environmental data includes data on various elements. The environmental data may include various data such as transfers, evaluations, salaries, overtime hours, marriage / divorce, and death of a close person. The environmental data may be data that can be acquired by a questionnaire for each person. The questionnaire may be conducted at each medical checkup or at any other time. The data stored in the attribute storage area 137 may be updated every time the server 1 acquires attribute data of any individual.

[0125] The auxiliary storage device 13 includes a factor prediction model storage area 138. The factor prediction model storage area 138 stores a factor prediction model. The disease risk prediction model is a model that predicts factors for mental illness based on attribute data of one or more individuals and time series data of mental illness risk of one or more individuals. For example, the attribute data of one or more individuals includes time series data of attribute data of one or more individuals. The disease risk prediction model outputs factor data for mental illness based on input of attribute data of one or more individuals and time series data of mental illness risk of one or more individuals. The factors for mental illness are factors related to mental illness. The factors for mental illness include disease prevention factors that are factors for the prevention of mental illness that reduce the possibility of suffering from mental illness. The factors for mental illness include disease aggravation factors that increase the possibility of suffering from mental illness. The factors for mental illness are elements included in the attribute data. The disease risk prediction model can be updated as appropriate.

[0126] Each unit realized by the above-mentioned processor 11 will be described. The processor 11 realizes a communication processing unit 110, a model processing unit 115, a first prediction processing unit 116, and a second prediction processing unit 117. Each unit realized in the processor 11 can also be referred to as each function. Each unit realized in the processor 11 can also be referred to as being realized in a control unit including the processor 11 and the main memory 12.

[0127] The communication processing unit 110 processes communication between the server 1 and the medical institution terminal 2 via the network NW using the communication interface 14. The communication processing unit 110 processes communication between the server 1 and the corporate terminal 4 via the network NW using the communication interface 14.

[0128] The model processing unit 115 generates a disease risk prediction model by machine learning, as in the first embodiment.

[0129] The first prediction processing unit 116 predicts the user's risk of mental illness based on the user's fundus information, similar to the prediction processing unit 111 described in the first embodiment.

[0130] The second prediction processing unit 117 predicts factors related to mental illness based on attribute data of one or more individuals and time-series data on mental illness risk of one or more individuals.

[0131] (Example of operation) The processing in the disease risk assessment support system 200 will be described. The process steps described below are merely examples, and each process may be modified as much as possible. In addition, steps may be omitted, replaced, or added as appropriate to the process steps described below depending on the embodiment.

[0132] FIG. 16 is a diagram illustrating an example of a flow of factors regarding mental illness in the disease risk assessment support system 200.

[0133] The medical institution terminal 2 transmits the fundus information of the user to the server 1 via the network NW. For example, every time the medical institution acquires the fundus information of the user, the medical institution terminal 2 transmits the fundus information of the user to the server 1. The server 1 receives the fundus information of the user from the medical institution terminal 2 via the network NW. The medical institution terminal 2 transmits the user's attribute data to the server 1 via the network NW. For example, every time the medical institution acquires the user's attribute data, the medical institution terminal 2 transmits the user's attribute data to the server 1. The user's attribute data includes data based on values ​​measurable from a living body. The value-based data may be data obtained by converting a measured value. The user's attribute data may include data obtained by converting an age into increments of 10 years or the like. The medical institution terminal 2 transmits such user's attribute data to the server 1 in order to anonymize the user's attribute data so that an individual cannot be identified even if multiple values ​​included in the user's attribute data are linked. The server 1 receives the user's attribute data from the medical institution terminal 2 via the network NW. The server 1 predicts the user's mental illness risk based on the user's fundus information. The server 1 predicts factors related to mental illness based on attribute data of one or more individuals including a user and time-series data of mental illness risk of one or more individuals. The server 1 transmits information indicating factors related to mental illness to the corporate terminal 4 via the network NW.

[0134] FIG. 17 is a flowchart illustrating a process related to disease preventive factors in the server 1.

[0135] The communication processing unit 110 receives the fundus information of the user from the medical institution terminal 2 using the communication interface 14 (step S61). The communication processing unit 110 stores the received fundus information of the user in the fundus information storage area 130 (step S62).

[0136] The communication processing unit 110 receives the user attribute data from the medical institution terminal 2 using the communication interface 14 (step S63). The communication processing unit 110 stores the received user attribute data in the attribute storage area 137 (step S64).

[0137] The first prediction processing unit 116 acquires the user's fundus information from the fundus information storage area 130 (step S65). The first prediction processing unit 116 acquires the user's fundus information when it is time to predict the mental illness risk. The time to predict the mental illness risk is any timing after the communication processing unit 110 receives the user's fundus information from the medical institution terminal 2.

[0138] The first prediction processing unit 116 predicts the user's mental illness risk based on the user's fundus information (step S66). In step S66, for example, the first prediction processing unit 116 inputs the user's fundus information to a disease risk prediction model stored in the disease risk prediction model storage area 132. The first prediction processing unit 116 acquires the user's mental illness risk data from the disease risk prediction model based on the input of the user's fundus information to the disease risk prediction model. The first prediction processing unit 116 stores the mental illness risk data of the user in the illness risk storage area 131 (step S67).

[0139] The second prediction processing unit 117 acquires attribute data of one or more individuals including the user from the attribute storage area 137 (step S68). The attribute data of one or more individuals includes time-series data of the attribute data of one or more individuals. The second prediction processing unit 117 acquires time series data of mental illness risk of one or more individuals including the user from the illness risk storage area 131 (step S69). The time series data of the attribute data of one or more individuals is data for the same period as the time series data of the attribute data of one or more individuals.

[0140] The second prediction processing unit 117 predicts factors for mental illness based on attribute data of one or more individuals and time series data of mental illness risk of one or more individuals (step S70). In step S70, for example, the second prediction processing unit 117 inputs attribute data of one or more individuals and time series data of mental illness risk of one or more individuals to a factor prediction model stored in the factor prediction model storage area 138. The second prediction processing unit 117 acquires factor data for mental illness from the factor prediction model based on the input of attribute data of one or more individuals and time series data of mental illness risk of one or more individuals to the factor prediction model. The second prediction processing unit 117 estimates the relationship of the attribute data to the time series data of mental illness risk by machine learning. The second prediction processing unit 117 predicts factors for mental illness based on the estimation. The machine learning is unsupervised learning. The machine learning is, but is not limited to, a neural network or the like.

[0141] The communication processing unit 110 transmits information indicating factors of mental illness to the company terminal 4 using the communication interface 14 (step S71). The communication processing unit 110 may transmit information indicating factors of mental illness to the medical institution terminal 2 using the communication interface 14. In this example, the medical institution terminal 2 transmits information indicating factors of mental illness to an individual's terminal. The medical institution terminal 2 may transmit information indicating factors of mental illness to each terminal of one or more individuals. The medical institution terminal 2 may transmit information indicating factors of mental illness to a terminal of an individual whose mental illness risk level is at or above a predetermined level. The communication processing unit 110 may transmit information indicating factors of mental illness to a terminal of an individual whose mental illness risk level is at or above a predetermined level using the communication interface 14. In this example, the communication processing unit 110 may transmit information indicating factors of mental illness to each terminal of one or more individuals using the communication interface 14. The communication processing unit 110 may transmit information indicating factors of mental illness to a terminal of an individual whose mental illness risk level is at or above a predetermined level.

[0142] The communication processing unit 110 transmits countermeasure information regarding countermeasures against mental illness risk to the medical institution terminal 2 based on the factors of the mental illness using the communication interface 14 (step S72). The countermeasure information is information indicating countermeasures suitable for avoiding and reducing the risk of mental illness. The countermeasure information may include information indicating products suitable for avoiding and reducing the risk of mental illness as countermeasures. The countermeasure information may include information indicating lifestyle habits suitable for avoiding and reducing the risk of mental illness as countermeasures. The countermeasure information may differ depending on the factors of the mental illness. This is because the appropriate countermeasure changes depending on the factors. If the factor is sleep time, the countermeasure may be improving lifestyle habits to ensure sleep time. If the factor is blood pressure, the countermeasure may be improving lifestyle habits or recommending a medical examination to bring blood pressure into the normal range. The auxiliary storage device 13 stores countermeasure information according to the factors of the mental illness. The medical institution terminal 2 transmits the countermeasure information to the individual's terminal. The medical institution terminal 2 may transmit the countermeasure information to each terminal of one or more individuals. The medical institution terminal 2 may transmit the countermeasure information to a terminal of an individual whose mental illness risk level is equal to or higher than a predetermined level. The communication processing unit 110 may transmit the countermeasure information to the individual's terminal using the communication interface 14. In this example, the communication processing unit 110 may transmit the countermeasure information to each terminal of one or more individuals using the communication interface 14. The communication processing unit 110 may transmit the countermeasure information to the terminal of an individual whose mental illness risk level is equal to or higher than a predetermined level.

[0143] (Modification) The server 1 may be an electronic device that can be used by a medical institution, similar to the medical institution terminal 2. In this example, since the medical institution can use the server 1 and the medical institution terminal 2, the medical institution terminal 2 does not need to perform processing to anonymize the attribute data of the user.

[0144] (effect) The health risk assessment support device includes a first prediction processing unit that predicts a health risk of a user based on fundus information of the user. The health risk assessment support device includes a second prediction processing unit that predicts factors related to health based on attribute data of one or more individuals including the user and time-series data of health risks of the one or more individuals related to health. The health risk assessment support device can predict health factors, such as diseases or mental health, that cause changes in health risks over time by continuously monitoring health risks. The health risk assessment support device can support the consideration of health measures by predicting health factors, and can therefore support health maintenance activities by individuals or organizations.

[0145] The attribute data includes at least one of data on lifestyle habits, biometric data, and environmental data. The health risk assessment support device can improve the accuracy of predicting health factors by using at least one of data on lifestyle habits closely related to health, biological data, and environmental data.

[0146] The health risk assessment support device transmits countermeasure information regarding countermeasures against health risks based on health factors. The health risk assessment support device can support health maintenance activities for avoiding and reducing health risks by providing countermeasure information based on health factors.

[0147] [Third embodiment] Hereinafter, the third embodiment will be described in detail with reference to the drawings. The same components as those in the first embodiment are denoted by the same reference numerals as much as possible, and duplicated explanations will be omitted. The third embodiment adds a function of evaluating the risk of mental illness to the disease risk assessment support device.

[0148] (Configuration example) The configuration of the disease risk assessment support system 100 may be the same as that of the first embodiment, and therefore the description thereof will be omitted. The configuration of the server 1 may be the same as that of the first embodiment, and therefore the description thereof will be omitted.

[0149] (Example of operation) The processing in the disease risk assessment support system 100 will be described. The process steps described below are merely examples, and each process may be modified as much as possible. In addition, steps may be omitted, replaced, or added as appropriate to the process steps described below depending on the embodiment.

[0150] The process of adding a function to evaluate mental illness risk will be described. Note that "adding a function to evaluate mental illness risk" is intended to add mental illness risk as a new evaluation target. Therefore, the expression "mental illness risk" can also be expressed as "new mental illness risk" which intends the mental illness risk to be a new evaluation target.

[0151] FIG. 19 is a flowchart illustrating a process for adding a function for evaluating the risk of a mental illness in the server 1. The model processing unit 115 may start processing to add a function for evaluating the risk of mental illness.

[0152] The model processing unit 115 confirms the correlation between the fundus information of a patient and the disease through an observational study of a patient whose risk of mental illness has developed (step S81). In step S81, for example, similar to the above-mentioned step S42, the model processing unit 115 estimates the relationship between the fundus information and data on the morbidity of a mental illness through machine learning based on the learning data for disease risk prediction, and generates a disease risk prediction model. The generation of a disease risk prediction model by the model processing unit 115 in this manner is an example of confirming the correlation between the fundus information of a patient and the disease through an observational study of a patient whose risk of mental illness has developed. The multiple learning subjects in the learning data for disease risk prediction are multiple patients whose risk of mental illness has developed, such as a small group of patients whose risk of mental illness has been confirmed. The fundus information of each of the multiple learning subjects included in the learning data for disease risk prediction is the fundus information of the patient obtained by an observational study of the patient using an OCT device or a fundus camera, etc.

[0153] The model processing unit 115 adds a function for evaluating mental illness risk to the server 1 based on the confirmation of correlation (step S82). If the accuracy of the function for evaluating mental illness risk is equal to or greater than the reference value, the model processing unit 115 adds a function for evaluating mental illness risk to the server 1 based on the confirmation of correlation. Adding the function for evaluating mental illness risk includes making the function for evaluating mental illness risk executable in the server 1. In step S82, for example, the model processing unit 115 stores the generated disease risk prediction model in the disease risk prediction model storage area 132, and makes the disease risk prediction model executable in the server 1.

[0154] The prediction processing unit 111 evaluates the user's mental illness risk based on the user's fundus information by using the function of evaluating the mental illness risk (step S83). In step S83, for example, similar to the above-mentioned step S4, the prediction processing unit 111 predicts the user's mental illness risk based on the user's fundus information.

[0155] When the user's mental illness risk is equal to or higher than a predetermined level, the model processing unit 115 presents the user with a question for verifying whether the mental illness risk is correct (step S84). For example, when the user's mental illness risk is equal to or higher than a predetermined level, the evaluation value indicated by the user's mental illness risk is equal to or higher than a predetermined value. The predetermined value can be set appropriately. When the user's mental illness risk is equal to or higher than a predetermined level, the user is in a state of high mental illness risk. In step S84, for example, the model processing unit 115 transmits a question to the user's terminal. The user only needs to be able to view the question, and the manner in which the question is presented to the user is not limited. The question is a question about the mental illness risk for verifying whether the evaluation value indicated by the mental illness risk is correct. For example, the question is a standard interview for quickly screening for a mental illness. When the mental illness is depression and the evaluation value indicated by the mental illness risk is a value of PHQ-9, the question is a question in the PHQ-9. In addition, when the server 1 evaluates the mental health risk instead of evaluating the mental illness risk, the evaluation value indicated by the risk may be the value of PHQ-9. In this example, the questions are questions in the PHQ-9. If the server 1 evaluates the disease risk of dementia (e.g., mild cognitive impairment (MCI)) instead of evaluating the mental disease risk, the questions are questions in the Mini Mental State Examination (MMSE).

[0156] Based on the provision of the question, the model processing unit 115 collects the user's answer to the question (step S85). In step 85, for example, the model processing unit 115 receives the user's answer transmitted from the user's terminal.

[0157] The model processing unit 115 executes machine learning based on learning data including the user's health risk and correct / incorrect answers (step S86). Correct / incorrect answers are data indicating whether the user's mental illness risk is correct or incorrect based on the user's answers. For example, correct / incorrect answers are data indicating whether the evaluation value indicated by the user's mental illness risk is correct (matches) or incorrect (mismatches) with respect to the value of the index of mental illness risk based on the user's answers. Correct / incorrect answers are an example of data based on the user's answers. The learning data may include data based on the user's answers. The data based on the user's answers may be the value of the index of mental illness risk based on the user's answers.

[0158] The model processing unit 115 updates the function for evaluating the mental illness risk added to the server 1 based on the execution of the machine learning (step S87). In step S87, for example, the model processing unit 115 updates the disease risk prediction model based on the execution of the machine learning. The model processing unit 115 makes the updated disease risk prediction model executable in the server 1.

[0159] Through the above-described processing, the server 1 can improve the accuracy of the function of assessing the risk of mental illness while executing the function of assessing the risk of mental illness.

[0160] Although an example in which the server 1 executes the multiple processes illustrated in FIG. 19 has been described, the present invention is not limited to this. The multiple processes illustrated in FIG. 19 may be executed by multiple devices including the server 1. For example, the process of step S83 described with the prediction processing unit 111 as the execution subject may be executed by the server 1. The multiple processes of steps S81, S82, S84, S85, S86, and S87 described with the model processing unit 115 as the execution subject may be executed by one other server different from the server 1. The multiple processes described with the model processing unit 115 as the execution subject may be executed by multiple other servers. In this case, in step S82, adding a function for evaluating a mental illness risk to the server 1 by the other server may include transmitting a disease risk prediction model to the server 1. In step S87, updating the function for evaluating a mental illness risk added to the server 1 by the other server may include transmitting the updated disease risk prediction model to the server 1.

[0161] [Other embodiments] Although an example of predicting mental illness risk based on fundus information has been described, the present invention is not limited to this. The prediction target based on fundus information may be a disease risk related to a disease that can be predicted from fundus information. In addition to mental illness, the disease may be dementia, glaucoma, diabetes, Alzheimer's, etc., but is not limited to these. In this example, the notation "mental illness" may be read as "disease". The notation "mental illness risk" may be read as "disease risk". The disease risk includes the risk of already suffering from a disease at the present time, as with mental illness risk. In this example, the notation "mental illness risk" may be read as "current disease risk". The disease risk includes the risk of not suffering from a disease at the present time, as with mental illness risk, at the present time, of suffering from a disease in the future. In this example, the notation "mental illness risk" may be read as "future disease risk". The prediction target based on fundus information may be a mental health state that can be predicted from fundus information. In this example, the notation "mental illness" or "affliction with mental illness" may be read as "mental health" or "mental health state". The term "mental illness risk" may be read as "mental health state". The mental health state includes the current mental health state, similar to mental illness risk. In this example, the term "mental illness risk" may be read as "current mental health state". The mental health state includes the future mental health state, similar to mental illness risk. In this example, the term "mental illness risk" may be read as "future mental health state". The prediction target based on fundus information may be a health risk that can be predicted from fundus information. In this example, the term "mental illness" or "mental illness morbidity" may be read as "health" or "health risk". The term "mental illness risk" may be read as "health risk". The health risk includes the current health risk, similar to mental illness risk. In this example, the term "mental illness risk" may be read as "current health risk". The health risk includes the future health risk, similar to mental illness risk. In this example, the term "mental illness risk" may be read as "future health risk".

[0162] Biometric data relating to the eyes may be used to predict the risk of mental illness. The biometric data relating to the eyes is data that can be measured from the eyes. The biometric data relating to the eyes may be fundus information or data based on values ​​that can be measured from the eyes. The biometric data relating to the eyes may be both fundus information and data based on values ​​that can be measured from the eyes. In this example, the term "fundus information" may be read as "biometric data relating to the eyes."

[0163] A disease risk for a disease may be predicted based on biometric data related to the eye. As described above, the disease may be, in addition to mental illness, glaucoma, diabetes, Alzheimer's, etc., but is not limited to these. In this example, the notation "fundus information" may be read as "biometric data related to the eye". The notation "mental illness" may be read as "disease". The notation "risk of mental illness" may be read as "risk of disease". The notation "risk of mental illness" may be read as "current risk of disease". The notation "risk of mental illness" may be read as "future risk of disease".

[0164] A disease risk for a disease may be predicted based on the biometric data. In this example, the biometric data may be biometric data related to the eyes, biometric data related to a part of the body other than the eyes, blood data, or tissue data of the body. The biometric data may be a combination of these. As described above, the disease may be, but is not limited to, glaucoma, diabetes, or Alzheimer's, in addition to mental diseases. The disease may be various diseases that can be predicted based on the biometric data. In this example, the notation "fundus information" may be read as "biological data". The notation "mental disease" may be read as "disease". The notation "mental disease risk" may be read as "disease risk". The notation "mental disease risk" may be read as "current disease risk". The notation "mental disease risk" may be read as "future disease risk". A mental health state may be predicted based on the biometric data. In this example, the notation "fundus information" may be read as "biological data". The notation "mental disease" or "mental disease" may be read as "mental health" or "mental health state". The term "mental illness risk" may be read as "mental health state." The term "mental illness risk" may be read as "current mental health state." The term "mental illness risk" may be read as "future mental health state." Health risks related to health may be predicted based on biometric data. In this example, the term "fundus information" may be read as "biometric data." The terms "mental illness" or "mental illness morbidity" may be read as "health" or "health risk." The term "mental illness risk" may be read as "health risk." The term "mental illness risk" may be read as "current health risk." The term "mental illness risk" may be read as "future health risk."

[0165] The health risk assessment support device may be realized by one electronic device such as the server 1, or may be realized by multiple electronic devices with the functions of the server 1 distributed. The health risk assessment support device may be realized by multiple electronic devices with the functions distributed, such as the server 1 and the medical institution terminal 2 connected to each other via a network NW. In this case, the health risk assessment support device includes the meaning of a health risk assessment support system realized by multiple electronic devices. When either the server 1 or the medical institution terminal 2 is an example of a first device realized by one or more electronic devices, the other is an example of a second device realized by one or more electronic devices.

[0166] The processing circuitry includes one or more circuits that realize multiple processes by multiple functions, for example, but not limited to, a processor, an ASIC (Application Specific Integrated Circuit), or an FPGA (Field-Programmable Gate Array).

[0167] Each of the one or more circuits constituting the processing circuit executes one or more processes among the multiple processes. When the processing circuit is composed of a single circuit, the single circuit executes all of the multiple processes. When the processing circuit is composed of multiple circuits, each of the multiple circuits executes a part of the multiple processes. The part of the multiple processes may be one of the multiple processes, or two or more of the multiple processes. When the processing circuit is composed of multiple circuits, the multiple circuits may be included in one device, or may be distributed across multiple devices.

[0168] The program may be transferred in a state where it is stored in an electronic device, or in a state where it is not stored in an electronic device. In the latter case, the program may be transferred via a network, or in a state where it is recorded on a recording medium. The recording medium is a non-transitory tangible medium. The recording medium is a computer-readable medium. The recording medium may be in any form, such as a CD-ROM or a memory card, as long as it is capable of storing the program and is readable by a computer.

[0169] Although the embodiment of the present invention has been described in detail above, the above description is merely an example of the present invention in every respect. It goes without saying that various improvements and modifications can be made without departing from the scope of the present invention. In other words, in implementing the present invention, a specific configuration according to the embodiment may be appropriately adopted.

[0170] In short, this invention is not limited to the above-mentioned embodiment as it is, and in the implementation stage, the components can be modified and embodied without departing from the gist of the invention. In addition, various inventions can be formed by appropriately combining multiple components disclosed in the above-mentioned embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, components from different embodiments may be appropriately combined.

[0171] (Additional Note) The embodiment can be expressed as follows. (1) a prediction processing unit that predicts a health risk of a user based on fundus information of the user; A communication processing unit that transmits information regarding a health risk of a subject to be evaluated, including the user; A health risk assessment support device comprising: (2) the prediction processing unit predicts a progression risk of a health risk of the user based on time-series data of fundus information of the user; The communication processing unit transmits information regarding the progression risk of the evaluation subject. A health risk assessment support device as described in (1). (3) A calculation unit is further provided for calculating a deviation of a health risk of the evaluation object based on the health risk data of a plurality of evaluation units stored in a storage unit, The information regarding the health risk of the subject to be evaluated includes information indicating a deviation of the health risk of the subject to be evaluated. A health risk assessment support device according to (1) or (2). (4) the evaluation target is an organization to which the user belongs, The set of the multiple evaluation units includes the organization to which the user belongs and is larger than the organization to which the user belongs, The deviation of the health risk of the evaluation object is the deviation of the health risk of the evaluation object for the set of the multiple evaluation units; A health risk assessment support device as described in (3). (5) the evaluation target is the individual user, the set of the plurality of evaluation units is an organization to which the user belongs, The deviation of the health risk of the evaluation object is the deviation of the health risk of the evaluation object for the set of the multiple evaluation units; A health risk assessment support device as described in (3). (6) The health risk assessment support device according to any one of (1) to (5), wherein the communication processing unit transmits countermeasure information regarding countermeasures against the health risk of the evaluation target based on the health risk of the evaluation target. (7) The health risk assessment support device according to (6), wherein the countermeasure information differs depending on the level of the health risk of the subject to assessment. (8) further comprising an evaluation processing unit that evaluates an effect of the countermeasure applied to the evaluation object on the health risk of the evaluation object based on time-series data of the health risk of the evaluation object. A health risk assessment support device according to (6) or (7). (9) The communication processing unit transmits guidance information related to a health risk of the evaluation subject. A health risk assessment support device according to any one of (1) to (8). (10) The guidance information varies depending on the level of the health risk of the subject to be evaluated. A health risk assessment support device as described in (9). (11) The guidance information differs depending on whether the guidance information is to be notified to the user or to an organization to which the user belongs. A health risk assessment support device as described in (9). (12) A health risk assessment support device described in any of (1) to (11), wherein the communications processing unit transmits insurance information regarding insurance that provides indemnification for the subject to be assessed for incapacity to work based on the health risk of the subject to be assessed. (13) A health risk assessment support device according to (12), wherein at least one of the insurance content and the insurance premium varies depending on the level of health risk of the subject to be assessed. (14) a first prediction processing unit that predicts a health risk of a user based on fundus information of the user; A second prediction processing unit that predicts factors related to the health based on attribute data of one or more individuals including the user and time-series data of the health risk of the one or more individuals related to health; A health risk assessment support device comprising: (15) The attribute data includes at least one of data on lifestyle habits, biometric data, and environmental data. A health risk assessment support device as described in (14). (16) A health risk assessment support device according to (14) or (15), which transmits countermeasure information regarding countermeasures against the health risk based on the health factors. (17) A program for enabling each unit of the health risk assessment support device described in any one of (1) to (16) to be executed. (18) A method for adding a function for evaluating a new health risk to a health risk assessment support device, comprising: (a) confirming a correlation between fundus information of the patient and the disease through an observational study of the patient who has developed the new health risk; (b) adding a function for evaluating the new health risk to the health risk assessment support device; (c) assessing the new health risk of the user based on the fundus information of the user; (d) if the new health risk of the user is equal to or greater than a predetermined value, presenting the user with a question to verify the accuracy of the new health risk; and (e) performing machine learning based on training data including the new health risk of the user and the correct or incorrect result; A method for adding a function for evaluating new health risks to a health risk assessment support device comprising: [Explanation of symbols]

[0172] 1...server, 2...medical institution terminal, 3...terminal for evaluation subject, 3-1...personal terminal, 3-2...organization terminal, 4...corporate terminal, 11...processor, 12...main memory, 13...auxiliary storage device, 14...communication interface, 100...disease risk assessment support system, 110...communication processing unit, 111...prediction processing unit, 112...calculation unit, 113...guidance processing unit, 114...evaluation processing unit, 115...model processing unit, 116...first prediction processing unit, 117...second prediction processing unit, 130...fundus information storage area, 131...disease risk storage area, 132...disease risk prediction model storage area, 133...learning data storage area for disease risk prediction, 134...progression risk prediction model storage area, 135...learning data storage area for progression risk prediction, 136...guidance storage area, 137...attribute storage area, 138...factor prediction model storage area, 200...disease risk assessment support system, NW...network.

Claims

1. A prediction processing unit that predicts a health risk of a user based on fundus information of the user; A communication processing unit that transmits information regarding a health risk of a subject to be evaluated, including the user; A health risk assessment support device comprising:

2. The prediction processing unit predicts a progression risk of a health risk of the user based on time-series data of fundus information of the user, The communication processing unit transmits information regarding the progression risk of the evaluation subject. The health risk assessment support device according to claim 1 .

3. A calculation unit is further provided for calculating a deviation of a health risk of the evaluation object based on the health risk data of a plurality of evaluation units stored in a storage unit, The information regarding the health risk of the subject to be evaluated includes information indicating a deviation of the health risk of the subject to be evaluated. The health risk assessment support device according to claim 1 .

4. the evaluation target is an organization to which the user belongs, The set of the multiple evaluation units includes the organization to which the user belongs and is larger than the organization to which the user belongs, The deviation of the health risk of the evaluation object is the deviation of the health risk of the evaluation object for the set of the plurality of evaluation units; The health risk assessment support device according to claim 3.

5. the evaluation target is the individual user, the set of the plurality of evaluation units is an organization to which the user belongs, The deviation of the health risk of the evaluation object is the deviation of the health risk of the evaluation object for the set of the multiple evaluation units; The health risk assessment support device according to claim 3.

6. The health risk assessment support device according to claim 1 , wherein the communication processing unit transmits countermeasure information regarding countermeasures against the health risk of the evaluation target, based on the health risk of the evaluation target.

7. The health risk assessment support device according to claim 6 , wherein the countermeasure information differs depending on a level of the health risk of the assessment subject.

8. and an evaluation processing unit that evaluates an effect of the countermeasure applied to the evaluation object on the health risk of the evaluation object based on time-series data of the health risk of the evaluation object. The health risk assessment support device according to claim 6 or 7.

9. The communication processing unit transmits guidance information related to a health risk of the evaluation subject. The health risk assessment support device according to claim 1 .

10. The guidance information varies depending on the level of health risk of the subject to be evaluated. The health risk assessment support device according to claim 9.

11. The guidance information differs depending on whether the guidance information is to be notified to the user or to an organization to which the user belongs. The health risk assessment support device according to claim 9.

12. 2. The health risk assessment support device according to claim 1, wherein the communication processing unit transmits insurance information relating to insurance that insures the subject against incapacity to work based on the health risk of the subject.

13. 13. The health risk assessment support device according to claim 12, wherein at least one of the insurance content and the insurance premium varies depending on the level of health risk of the subject to be assessed.

14. A first prediction processing unit that predicts a health risk of a user based on fundus information of the user; A second prediction processing unit that predicts factors related to the health based on attribute data of one or more individuals including the user and time-series data of the health risk of the one or more individuals related to health; A health risk assessment support device comprising:

15. The attribute data includes at least one of data on lifestyle habits, biometric data, and environmental data. The health risk assessment support device according to claim 14.

16. 16. The health risk assessment assistance device according to claim 14, further comprising: transmitting countermeasure information regarding countermeasures against the health risk based on the health factors.

17. A program for enabling each unit of the health risk assessment support device according to any one of claims 1 to 16 to be executed.

Images