Disease examination intervention recommendation method, disease examination intervention recommendation device, and disease examination intervention recommendation program

Abstract

To recommend an appropriate intervention method according to a cause of a disease risk predicted for each person.SOLUTION: A disease examination intervention recommendation method is executed by a computer system having a processor and a storage device. The storage device holds health information indicating a health state of a person and correspondence information for associating a disease risk factor and an intervention method with each other. The disease examination intervention recommendation method includes: a procedure of inputting, by the processor, information on a plurality of items indicating a health state of a prediction target person as an explanatory variable, to a disease risk prediction model that predicts a disease risk for a person on the basis of information on a plurality of items indicating a health state of the person, thereby predicting the disease risk for the prediction target person and calculating the degree of contribution of the items of the explanatory variable for a result of prediction of the disease risk for the prediction target person; and a procedure of selecting, by the processor, the intervention method on the basis of the disease risk factor corresponding to the input explanatory variable, the magnitude of the calculated degree of contribution, and the correspondence information.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to technology that assists in recommending tests and interventions for diseases such as dementia. [Background technology]

[0002] With regard to a technique for improving a doctor's workflow regarding diseases, for example, the technique disclosed in International Publication No. 2013 / 144803 (Patent Document 1) is known.

[0003] Patent document 1 describes a system for improving workflow that includes one or more clinical data sources that collect patient data from patients, a patient information system that stores the patient data, and a clinical decision support system that includes one or more processors that are programmed to receive patient data from patients, generate quantified information for each type of patient data based on a statistical model, diagnose the patient based on the quantified information, generate recommendations based on the diagnosis and the quantified information, and display the recommendations. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] International Publication No. 2013 / 144803 Summary of the Invention [Problem to be solved by the invention]

[0005] Regarding tests and interventions to address disease risk, even if symptoms of disease are noticed, it is difficult to know the appropriate next test or intervention method. Taking dementia as an example, the causes of dementia risk may differ from person to person, and in order to select the appropriate next intervention method, it is necessary to analyze the causes of dementia risk. [Means for solving the problem]

[0006] In order to solve at least one of the above problems, the present invention provides a disease testing intervention recommendation method executed by a computer system having a processor and a storage device, wherein the storage device holds health information indicating a person's health condition and correspondence information that associates disease risk factors with intervention methods, and the disease testing intervention recommendation method is characterized by including: a first step in which the processor predicts the disease risk of the person to be predicted by inputting information on multiple items that indicate the health condition of the person as explanatory variables into a disease risk prediction model that predicts the disease risk of the person based on information on multiple items that indicate the person's health condition, and calculates the contribution of each item of the explanatory variables to the prediction result of the disease risk of the person to be predicted; and a second step in which the processor identifies the disease risk factors that correspond to the input explanatory variables, and selects an intervention method based on the identified disease risk factors, the magnitude of the calculated contribution, and the correspondence information. [Effects of the Invention]

[0007] According to one aspect of the present invention, it is possible to recommend an appropriate next intervention method for each subject depending on the cause of the predicted disease risk.

[0008] Other problems, configurations and effects will become apparent from the following description of the embodiments. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a block diagram showing an example of the configuration of a dementia test intervention recommendation system according to an embodiment of the present invention. [Figure 2] FIG. 1 is a block diagram showing an example of a hardware configuration for realizing a dementia test intervention recommendation system according to an embodiment of the present invention. [Figure 3] FIG. 2 is an explanatory diagram showing an example of user information included in a user information DB of the dementia test intervention recommendation system in an embodiment of the present invention. [Figure 4]FIG. 1 is an explanatory diagram showing an example of digital biomarker information included in a health information DB of a dementia test intervention recommendation system in an embodiment of the present invention. [Figure 5] FIG. 2 is an explanatory diagram showing a first example of KDB information included in the health information DB of the dementia test intervention recommendation system in an embodiment of the present invention. [Figure 6] FIG. 10 is an explanatory diagram showing a second example of KDB information included in the health information DB of the dementia test intervention recommendation system in an embodiment of the present invention. [Figure 7] FIG. 10 is an explanatory diagram showing a third example of KDB information included in the health information DB of the dementia test intervention recommendation system in an embodiment of the present invention. [Figure 8] FIG. 10 is an explanatory diagram showing an example of a correspondence table between risk factors and dietary interventions included in the correspondence table of the dementia testing intervention recommendation system in an embodiment of the present invention. [Figure 9] FIG. 10 is an explanatory diagram showing an example of a correspondence table between risk factors and exercise interventions included in the correspondence table of the dementia testing intervention recommendation system in an embodiment of the present invention. [Figure 10] FIG. 10 is an explanatory diagram showing an example of a correspondence table between risk factors and social participatory interventions included in the correspondence table of the dementia testing intervention recommendation system in an embodiment of the present invention. [Figure 11] FIG. 10 is an explanatory diagram showing an example of a correspondence table between dementia risk probabilities and risk stages included in the correspondence table of the dementia testing intervention recommendation system in an embodiment of the present invention. [Figure 12] 1 is a flowchart illustrating an example of processing executed by the dementia test intervention recommendation system according to an embodiment of the present invention. [Figure 13] FIG. 10 is an explanatory diagram showing an example of information obtained by processing executed by the dementia test intervention recommendation system in an embodiment of the present invention. [Figure 14] FIG. 10 is an explanatory diagram showing an example of information output by the dementia test intervention recommendation system in an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, embodiments of the present invention will be described with reference to the drawings. Note that, although the following embodiments will be described with reference to dementia, which is an example of a disease, the present embodiments can be applied to any disease other than dementia. In other words, "dementia" in the following description can be replaced with "disease" or the name of a specific disease.

[0011] FIG. 1 is a block diagram showing an example of the configuration of a dementia test intervention recommendation system according to an embodiment of the present invention.

[0012] The dementia test intervention recommendation system 100 of this embodiment includes an input unit 101 , an output unit 102 , a calculation unit 103 , an extraction unit 104 and a storage unit 105 .

[0013] The input unit 101 accepts input of information from the administrator (not shown) of the dementia testing intervention recommendation system 100, the user 140, etc. The output unit 102 outputs information to the administrator of the dementia testing intervention recommendation system 100, the user 140, etc.

[0014] The calculation unit 103 has a dementia risk analysis processing unit 111. The extraction unit 104 has a test recommendation analysis processing unit 112, an intervention service recommendation analysis processing unit 113, and a local hospital recommendation analysis processing unit 114. The processing of each of these units will be described later.

[0015] The storage unit 105 holds a user information database (DB) 115, a health information DB 116, a correspondence table 117, a testing company DB 118, an intervention service provider DB 119, and a local hospital DB 120. Details of this information will be described later.

[0016] The dementia testing intervention recommendation system 100 is connected to a system of a user 140 and an external system 150 via a communication network 130. The system of the user 140 includes, for example, a smartphone 141 and a wearable device 142. The smartphone 141 is an information terminal device carried by the user 140, connected to the communication network 130, and communicating with the dementia testing intervention recommendation system 100 and the like. The wearable device 142 is a device that has the function of measuring the state of the user 140 and outputting data including the results of the measurement.

[0017] The wearable device 142 is, for example, a device worn by the user 140 to acquire the amount of activity, vital data, and the like of the user 140, and may be, for example, a so-called smart watch. The data acquired from the wearable device 142 depends on the model of the device, but may include, for example, at least one of the following: number of steps, walking distance, pulse rate, blood oxygen saturation, body temperature, sleep time, and conversation time. The wearable device 142 transmits the acquired data to the dementia testing intervention recommendation system 100 via the communication network 130. The dementia testing intervention recommendation system 100 stores this data in the health information DB 116 as digital biomarker (dBM) information. The wearable device 142 may be connected to the communication network 130 directly or via the smartphone 141. Furthermore, the health information DB 116 can collect information from a non-invasive sensor by asking the user 140, thereby solving the problem of cognitive function testing being time-consuming even if cognitive function testing tools are available.

[0018] The above-described system of user 140 is an example, and may include devices other than those described above. For example, a stationary personal computer (PC) may be used instead of (or in addition to) smartphone 141, and a stationary blood pressure monitor, weight scale, or the like capable of outputting measurement value data may be used instead of (or in addition to) wearable device 142.

[0019] Although FIG. 1 shows one user 140, in reality there are multiple users, each of whom has a smartphone 141 and a wearable device 142.

[0020] The external system 150 includes a testing company 151, an intervention service provider 152, and a local hospital 153. In practice, computer systems (not shown) used by the testing company 151, the intervention service provider 152, and the local hospital 153 for their business are connected to the communication network 130.

[0021] Testing company 151 is a company that provides dementia tests. The tests provided may be non-invasive tests such as task-based tests, or invasive tests such as blood tests. Although one testing company 151 is shown in FIG. 1, in reality there may be multiple testing companies 151, each of which may provide a different test.

[0022] The intervention service provider 152 is a provider that provides intervention services according to the risk of dementia. In this embodiment, multiple types of interventions are assumed according to the risk factors of dementia. For example, if the main risk factor is a dietary factor such as nutritional deficiency or imbalance, dietary interventions such as nutritional guidance, meal provision, and food delivery are considered effective. Also, if the main risk factor is an exercise factor such as lack of exercise, exercise interventions such as exercise guidance and introduction to exercise facilities are considered effective. Furthermore, if the main risk factor is a social participation factor such as social isolation, social participation interventions such as home visits and introduction to local clubs are considered effective.

[0023] The intervention service provider 152 is a provider that provides at least one of these types of intervention services. Although one intervention service provider 152 is shown in FIG. 1, in reality there may be multiple intervention service providers 152, each of which may provide a different intervention service. Furthermore, the intervention service provider 152 may be a private company or a public institution such as a local government.

[0024] The local hospital 153 is a hospital located in each region, and provides medical services such as dementia testing (especially detailed examinations by doctors), treatment, and hospitalization of dementia patients. The local hospital 153 may also function as part of the testing company 151 and the intervention service provider 152. While one local hospital 153 is shown in Figure 1, there may actually be multiple local hospitals 153, each providing different medical services.

[0025] The testing company DB 118 includes information such as the location, contact information, details of tests provided, and how to apply for tests of each testing company 151. Similarly, the intervention service provider DB 119 includes information such as the location, contact information, details of intervention services provided, target areas, and how to apply for each intervention service provider 152. Furthermore, the local hospital DB 120 includes information such as the location, contact information, details of medical services provided, and how to apply for a consultation of each local hospital 153.

[0026] FIG. 2 is a block diagram showing an example of a hardware configuration for realizing the dementia test intervention recommendation system 100 according to an embodiment of the present invention.

[0027] The dementia test intervention recommendation system 100 of this embodiment shown in Fig. 1 can be realized by a computer system. Fig. 2 shows a computer system 200 as an example.

[0028] The computer system 200 includes a processor 201, a memory (main storage device) 202, an auxiliary storage device 203, an output device 204, an input device 205, and a communication interface (I / F) 206. The above components are connected to each other via a bus. The memory 202 and the auxiliary storage device 203 are storage devices that store programs and data used by the processor 201. The memory 202 and the auxiliary storage device 203 correspond to the storage unit 105 in FIG. 1.

[0029] The memory 202 is configured, for example, by a semiconductor memory, and is mainly used to hold programs and data currently being executed. For example, programs and data stored in the auxiliary storage device 203 are loaded into the memory 202 at startup or when needed. The processor 201 executes various processes in accordance with the programs stored in the memory 202. The processor 201 operates in accordance with the programs to realize various functional units (for example, the calculation unit 103, extraction unit 104, input unit 101, and output unit 102 shown in FIG. 1).

[0030] The auxiliary storage device 203 is configured with a large-capacity storage device such as a hard disk drive or a solid state drive, and is used to store programs and data for a long period of time. For example, the programs and data may be stored in a user information DB 115, a health information DB 116, a correspondence table 117, a testing company DB 118, an intervention service provider DB 119, and a local hospital DB 120.

[0031] Processor 201 may be comprised of a single processing unit or multiple processing units and may include single or multiple arithmetic units or multiple processing cores. Processor 201 may be implemented as one or more central processing units, microprocessors, microcomputers, microcontrollers, digital signal processors, state machines, logic circuits, graphics processing units, systems on a chip, and / or any device that manipulates signals based on control instructions.

[0032] The input device 205 is a hardware device through which a user inputs instructions, information, etc. The output device 204 is a hardware device that presents various images for input and output, such as a display device such as a touch panel, a keyboard, or a printing device. The communication I / F 206 is an interface for connection to the communication network 130. For example, the functions of the input unit 101 and the output unit 102 are realized by a processor controlling at least one of the input device 205, the output device 204, and the communication I / F 206 in accordance with a program.

[0033] The computer system 200 may include two or more processors 201. Furthermore, the functions of the dementia testing intervention recommendation system 100 can be implemented in multiple computer systems 200. In this case, the multiple computer systems 200 communicate with each other via a communication network 130. For example, some of the multiple functions of the system of this embodiment may be implemented in one computer system 200, and other parts may be implemented in other computer systems 200.

[0034] Next, the user information DB 115 will be described with reference to FIG.

[0035] FIG. 3 is an explanatory diagram showing an example of user information included in the user information DB 115 of the dementia test intervention recommendation system 100 in the embodiment of the present invention.

[0036] 3 includes a user ID 301, gender 302, date of birth 303, age 304, address / zip code 305, and educational background 306. The user ID 301 is identification information for each user 140. The gender 302, date of birth 303, age 304, address / zip code 305, and educational background 306 are each user's gender, date of birth, age, residence code, and educational background (e.g., highest level of education, etc.), respectively.

[0037] Next, health information DB 116 will be described with reference to Figures 4 to 7. Health information DB 116 stores information related to the health condition of user 140. Any information related to the health condition may be used, but typical examples include so-called digital biomarker (dBM) information obtained from wearable device 142 carried by user 140 and information obtained from the National Health Insurance Database (KDB). The latter KDB information includes, for example, information on the results of health checkups taken by user 140 and records of visits to medical institutions by user 140.

[0038] FIG. 4 is an explanatory diagram showing an example of digital biomarker information included in the health information DB 116 of the dementia test intervention recommendation system 100 in the embodiment of the present invention.

[0039] The digital biomarker information 400 shown in FIG. 4 includes a user ID 401 , number of steps 402 , walking distance 403 , pulse rate 404 , blood oxygen saturation 405 , calories burned 406 , sleeping time 407 , and talking time 408 .

[0040] The user ID 401 is identification information for each user 140. The number of steps 402, walking distance 403, pulse rate 404, blood oxygen saturation level 405, calories burned 406, sleep time 407, and conversation time 408 are the number of steps, walking distance, pulse rate, blood oxygen saturation level, calories burned, sleep time, and conversation time, respectively, acquired from the wearable device 142 of each user 140. The registered values may be the total or average of the values measured by the wearable device 142 for a predetermined period of time, such as one day.

[0041] Note that the above information is an example, and the actual digital biomarker information 400 may not include at least one of the above items, or may include items other than those above (for example, body temperature, etc.).

[0042] FIG. 5 is an explanatory diagram showing a first example of KDB information included in the health information DB 116 of the dementia test intervention recommendation system 100 in the embodiment of the present invention.

[0043] The KDB information 500 shown in Fig. 5 includes information indicating the results of a medical checkup for each user. Specifically, the KDB information 500 includes a user ID 501 that identifies each user 140 and values of each item obtained from the results of the medical checkup taken by each user 140. In the example of Fig. 5, the items of values obtained from the results of the medical checkup include Body Mass Index (BMI) 502, abdominal circumference 503, systolic blood pressure 504, diastolic blood pressure 505, triglycerides 506, HDL cholesterol 507, LDL cholesterol 508, fasting blood glucose 509, HbA1c 510, urinary sugar 511, urinary protein 512, and serum albumin level 513. However, these are merely examples, and at least one of these items may be omitted, or other items may be further included.

[0044] FIG. 6 is an explanatory diagram showing a second example of KDB information included in the health information DB 116 of the dementia test intervention recommendation system 100 in the embodiment of the present invention.

[0045] The KDB information 600 shown in Fig. 6 includes medical institution consultation records (e.g., medical receipt information, etc.) of each user 140. For example, the KDB information 600 includes health checkup and health guidance information, medical information, and nursing care information of each user 140. In the example of Fig. 6, the KDB information 600 includes a user ID 601 that identifies each user 140, and disease information 602 to 605 of each user 140. The disease information 602 and 603 each indicate whether or not each user 140 has been diagnosed with a specific disease.

[0046] Similarly, disease information 604 and 605 indicate the diagnosis of dementia-related diseases. Disease information 604 indicates whether each user 140 has been diagnosed with a mental or behavioral disorder (ICD10 code: F00-F09). Disease information 605 indicates whether each user 140 has been diagnosed with another degenerative disease of the nervous system (ICD10 code: G30-G32).

[0047] Although FIG. 6 shows disease information 602 to 605, the KDB information 600 actually includes information on whether each user 140 has been diagnosed with each of a large number of diseases.

[0048] FIG. 7 is an explanatory diagram showing a third example of KDB information included in the health information DB 116 of the dementia test intervention recommendation system 100 in the embodiment of the present invention.

[0049] 7 shows the answers of each user 140 to a questionnaire for elderly people. The questionnaire may be created, for example, when the user 140 undergoes a health checkup, when the user 140 visits a medical institution, or on any other occasion.

[0050] Specifically, the KDB information 700 includes a user ID 701 for identifying each user and an elderly person questionnaire 702 indicating the content of the question. The example in Fig. 7 shows that one user 140 answered "Yes" to the question "Do you go out alone by bus or train?", another user 140 answered "No" to the question "Do you shop for daily necessities?", and yet another user answered "Yes" to the question "Do you give advice to family and friends?"

[0051] Next, correspondence table 117 will be described with reference to Figures 8 to 11. Correspondence table 117 includes information that associates risk factors with risk degrees indicating the magnitude of risk and intervention methods, as well as information that associates calculated dementia risk probabilities with dementia risk stages. The former information includes information on dietary risk factors, exercise risk factors, and social participation risk factors.

[0052] FIG. 8 is an explanatory diagram showing an example of a correspondence table between risk factors and dietary interventions included in the correspondence table 117 of the dementia testing intervention recommendation system 100 in the embodiment of the present invention.

[0053] The correspondence table 800 between risk factors and dietary interventions shown in Figure 8 is an example of information that associates dietary risk factors with dietary interventions, and specifically includes an intervention method 801, data 802, dietary risk factors 803, and risk frequency 804. The intervention method 801 is a "dietary intervention service." The dietary risk factors 803 indicate risk factors related to diet among the risk factors for dementia. Each value of the dietary risk factors 803 is assigned an identifier such as "d1" or "d2." The data 802 indicates the data from which each risk factor is obtained. The risk frequency 804 indicates the magnitude of dementia risk corresponding to each risk factor. In this example, the larger the value of the risk frequency 804, the greater the risk (the same applies to the risk frequency 904 in Figure 9 and the risk frequency 1004 in Figure 10, which will be described later).

[0054] For example, the first row of the correspondence table 800 between risk factors and dietary interventions shown in FIG. 8 indicates that the value of dietary risk factor 803, "BMI<20," is obtained from the results of a health checkup, and the value of its risk frequency 804 is "1." Meanwhile, the second row indicates that the value of dietary risk factor 803, "BMI<18.5," is obtained from the results of a health checkup, and the value of its risk frequency 804 is "2." These indicate that a BMI of less than 20 indicates a risk of dementia, that the risk increases further when a BMI of less than 18.5 is used, and that dietary intervention is considered effective in reducing the risk. The correspondence between risk factors, risk frequencies, and intervention methods is preset based on, for example, past dementia diagnoses and care results. The same applies to the other dietary risk factors, exercise risk factors, and social participation risk factors described below.

[0055] Similarly, in the example of Figure 8, dietary risk factors 803 obtained from health checkup information include weight loss rate, serum albumin level, etc. Dietary risk factors 803 obtained from responses to a standard questionnaire include whether or not the person can chew their food, etc. Dietary risk factors 803 obtained from responses to a questionnaire for elderly people include whether or not the person is eating properly, etc. The correspondence table 800 between risk factors and dietary interventions contains information that associates the value of each dietary risk factor 803 with the value of risk frequency 804.

[0056] FIG. 9 is an explanatory diagram showing an example of a correspondence table between risk factors and exercise interventions included in the correspondence table 117 of the dementia testing intervention recommendation system 100 in the embodiment of the present invention.

[0057] The correspondence table 900 between risk factors and exercise interventions shown in FIG. 9 is an example of information that associates exercise risk factors with exercise interventions, and specifically includes an intervention method 901, data 902, exercise risk factors 903, and risk degree 904. The intervention method 901 is an "exercise intervention service." The exercise risk factors 903 indicate exercise-related risk factors among the risk factors for dementia. Each value of the exercise risk factors 903 is assigned an identifier such as "e1" or "e2." The data 902 indicates the data from which each risk factor is obtained. The risk degree 904 indicates the magnitude of dementia risk corresponding to each risk factor.

[0058] For example, the fifth row of the correspondence table 900 between risk factors and exercise interventions shown in Figure 9 indicates that the value of the exercise risk factor 903, "BMI>25.0," is obtained from the results of a health check, and the value of its risk frequency 904 is "1." On the other hand, the sixth row indicates that the value of the exercise risk factor 903, "BMI>30.0," is obtained from the results of a health check, and the value of its risk frequency 904 is "2." These indicate that a BMI greater than 25.0 indicates a risk of dementia, that the risk increases further when the BMI is greater than 30.0, and that exercise intervention is considered effective in reducing the risk.

[0059] Similarly, in the example of FIG. 9, exercise risk factors 903 obtained from digital biomarker information include the number of steps and heart rate for each age and gender category. Furthermore, exercise risk factors 903 obtained from health checkup information include systolic blood pressure, diastolic blood pressure, LDL cholesterol, HDL cholesterol, etc. Exercise risk factors 903 obtained from responses to a standard questionnaire include the presence or absence of exercise habits. Exercise risk factors 903 obtained from responses to a questionnaire for elderly people include the level of motor function (e.g., walking speed, whether or not a fall has occurred). The correspondence table 900 between risk factors and exercise interventions includes information that associates the value of each exercise risk factor 903 with the value of risk score 904.

[0060] FIG. 10 is an explanatory diagram showing an example of a correspondence table between risk factors and social participatory interventions included in the correspondence table 117 of the dementia testing intervention recommendation system 100 in the embodiment of the present invention.

[0061] The correspondence table 1000 between risk factors and social participation interventions shown in FIG. 10 is an example of information that associates social participation risk factors with social participation interventions, and specifically includes an intervention method 1001, data 1002, social participation risk factors 1003, and risk degrees 1004. The intervention method 1001 is a "social participation intervention service." The social participation risk factors 1003 indicate risk factors related to social participation among the risk factors for dementia. Each value of the social participation risk factors 1003 is assigned an identifier such as "c1" or "c2." The data 1002 indicates the data from which each risk factor is obtained. The risk degrees 1004 indicate the magnitude of dementia risk corresponding to each risk factor.

[0062] For example, the first row of the correspondence table 1000 between risk factors and social participation interventions shown in Fig. 10 indicates that the value of the social participation risk factor 1003, "conversation time < 15 minutes / day (monthly average)", is acquired from digital biomarker information, and the value of its risk degree 1004 is "1." This indicates that if the average daily conversation time over a month is shorter than 15 minutes, there is a risk of dementia, and that social participation intervention is considered to be effective in reducing that risk.

[0063] Similarly, in the example of Fig. 10, social participation risk factors 1003 obtained from digital biomarker information include the number of times participating in various groups, the number of times meeting with friends and acquaintances, etc. Furthermore, social participation risk factors 1003 obtained from responses to a questionnaire for elderly people include the frequency of going out, whether or not one associates with others, etc. The correspondence table 1000 between risk factors and social participation interventions includes information that associates the value of each social participation risk factor 1003 with the value of risk frequency 1004.

[0064] FIG. 11 is an explanatory diagram showing an example of a correspondence table between dementia risk probabilities and risk stages included in the correspondence table 117 of the dementia test intervention recommendation system 100 in the embodiment of the present invention.

[0065] The correspondence table 1100 between dementia risk probability and risk stage shown in Figure 11 is an example of information that corresponds the dementia risk probability of each user 140 calculated based on the dementia risk prediction model, the dementia risk stage of the user 140, and the testing method, and specifically includes dementia risk probability 1101, dementia risk 1102, and testing method 1103.

[0066] Dementia risk probability 1101 indicates the range of values of dementia risk probability p for each user 140 calculated based on the dementia risk prediction model. Dementia risk 1102 indicates the level of dementia risk corresponding to each range of dementia risk probability p. Testing method 1103 indicates the appropriate testing method corresponding to each level.

[0067] In the example of Figure 11, if the dementia risk probability 1101 is in the range of 0 to 0.25, the dementia risk 1102 is determined to be the lowest, "healthy." Similarly, if the dementia risk probability 1101 is greater than 0.25 and less than or equal to 0.5, the dementia risk 1102 is determined to be "low risk," if the dementia risk probability 1101 is greater than 0.5 and less than or equal to 0.75, the dementia risk 1102 is determined to be "medium risk," and if the dementia risk probability 1101 is greater than 0.75 and less than or equal to 1.0, the dementia risk 1102 is determined to be "high risk."

[0068] If the dementia risk 1102 is "normal" or "low risk", the test method 1103 is "task-type test". If the dementia risk 1102 is "medium risk", the test method 1103 is a blood test. If the dementia risk 1102 is "high risk", the test method 1103 is a "specialist test".

[0069] The above-described correspondence between dementia risk probability and testing method is one example, and other correspondences may also be used. Generally, when the dementia risk is low, non-invasive tests that are low in cost and place a small burden on the user 140 are recommended, and as the dementia risk increases, invasive tests, specialist tests, etc. that provide more detailed information but increase in cost and burden on the user 140 are recommended.

[0070] Next, the processing executed by the dementia test intervention recommendation system 100 will be described with reference to FIGS. 12 and 13. FIG.

[0071] FIG. 12 is a flowchart showing an example of processing executed by the dementia test intervention recommendation system 100 in the embodiment of the present invention.

[0072] First, the dementia risk analysis processing unit 111 references the user information DB 115 and the health information DB 116 and generates, by machine learning, a model for predicting the dementia risk from information indicating the health condition of the user 140 (step 1201). Specifically, the dementia risk analysis processing unit 111 performs machine learning using values of items corresponding to the dementia disease history of each user 140, among the information included in the user information DB 115 and the health information DB 116, as objective variables and values of other items as explanatory variables, thereby generating a model for predicting the dementia risk (i.e., the risk of cognitive decline) of the user 140. Here, items corresponding to the dementia disease history are, for example, F00-F09 and G30-G32 of the ICD10 code, and it is particularly desirable to select F00, F01, F02, F03, F06.7, and G30 as objective variables.

[0073] For both the explanatory variables and the objective variables, binary data indicating whether the value of each item satisfies a predetermined condition may be used instead of using the value of each item as is. For example, when BMI is used as the explanatory variable, binary data indicating whether or not the value falls under BMI<20 (e.g., 1: yes, 0: no), or binary data indicating whether or not the serum albumin level falls under serum albumin≦3.5 g / dl (e.g., 1: yes, 0: no), etc. may be used as the explanatory variable.

[0074] Although the machine learning method is not limited, for example, methods such as logistic regression, decision tree, Light GBM, and XGBoost can be used. In addition, once the dementia risk prediction model is created, it can be continuously used thereafter. However, when at least one of the user information DB115 and the health information DB116 is updated, it is desirable to update the dementia risk prediction model by performing machine learning based on the updated data. For example, when the KDB information is updated every month, the dementia risk prediction model may be updated by performing machine learning in accordance with the update.

[0075] Next, the dementia risk analysis processing unit 111 calculates the dementia risk probability p of the user to be predicted by inputting the explanatory variables of the user 140 to be predicted into the generated dementia risk prediction model (steps 1202, 1203).

[0076] Then, the dementia risk analysis processing unit 111 determines the dementia risk of the user 140 to be predicted by comparing the calculated dementia risk probability p with the correspondence table 1100 between the dementia risk probability and the risk level (steps 1204 to  1207).

[0077] When the dementia risk probability p satisfies 0 ≦ p ≦ 0.25, the dementia risk analysis processing unit 111 determines that the user 140 to be predicted is healthy (step 1204). When the dementia risk probability p satisfies 0.25 < p ≦ 0.5, the dementia risk analysis processing unit 111 determines that the dementia risk of the user 140 to be predicted is low risk (step 1205). In these cases, the dementia risk analysis processing unit 111 determines that the test suitable for the user 140 to be predicted is a non-invasive task type test, and outputs the result to the extraction unit 104. With this configuration, the problem that the quantitative index regarding the cognitive function is not clear and the degree of decline in the cognitive function is difficult to understand is solved.

[0078] The inspection recommendation analysis processing unit 112 of the extraction unit 104 searches the inspection company DB 118, extracts information related to inspections, and outputs it to the user 140. For example, the inspection recommendation analysis processing unit 112 may extract one or more inspection companies 151 that provide inspection services suitable for the user 140 to be predicted, and output information related to those inspection companies 151, information for applying for inspections, etc. to the user 140.

[0079] When the determination result of the dementia risk is normal or low risk, in the example of FIG. 12, since a task type inspection is performed, the inspection recommendation analysis processing unit 112 extracts an inspection company 151 that provides a service for the task type inspection. When the user 140 needs to go to the place where the inspection is performed, an inspection company 151 that can provide a place close to the address of the user 140 as the place where the inspection is performed may be extracted. The user 140 undergoes a task type inspection according to the result (step 1208) and can obtain the inspection result (step 1212).

[0080] When the dementia risk probability p satisfies 0.5 < p ≤ 0.75, the dementia risk analysis processing unit 111 determines that the dementia risk of the user 140 to be predicted is medium risk (step 1206). In this case, the dementia risk analysis processing unit 111 determines that the inspection suitable for the user 140 to be predicted is an invasive blood test, and outputs the result to the extraction unit 104.

[0081] The inspection recommendation analysis processing unit 112 of the extraction unit 104 searches the inspection company DB 118, extracts information related to inspections, and outputs it to the user 140. This search can be performed in the same manner as when the dementia risk is normal or low risk. However, in the example of FIG. 12, since it is determined that a blood test is appropriate when the dementia risk is medium risk, the inspection recommendation analysis processing unit 112 extracts an inspection company 151 that provides a service for the blood test. The user 140 undergoes a blood test according to the result (step 1209) and can obtain the inspection result (step 1212). Here, the health information DB 116 may be updated according to the inspection result obtained here.

[0082] When the dementia risk probability p satisfies 0.75 < p ≤ 1.0, the dementia risk analysis processing unit 111 determines that the dementia risk of the user 140 to be predicted is a high risk (step 1207). In this case, the dementia risk analysis processing unit 111 determines that the examination suitable for the user 140 to be predicted is a more detailed specialist examination, and outputs the result to the extraction unit 104.

[0083] The regional hospital recommendation analysis processing unit 114 of the extraction unit 104 searches the regional hospital DB 120, extracts information on the regional hospital 153 where a specialist examination can be received, and outputs it to the user 140. At this time, the regional hospital 153 whose location is close to the address of the user 140 may be extracted. The user 140 undergoes a specialist examination according to the result (step 1210), and further necessary examinations, such as a blood test (step 1209), or an MMSE (Mini-Mental State Examination) or amyloid PET (Positron Emission Tomography) examination (step 1211) according to the judgment of the specialist, and can obtain the examination result (step 1212).

[0084] In step 1202, the dementia risk analysis processing unit 111 further calculates a risk coefficient a, which is the contribution degree of each explanatory variable to the predicted value of each explanatory variable when predicting the dementia risk by inputting each explanatory variable related to the user 140 to be predicted into the dementia risk prediction model. For example, the dementia risk analysis processing unit 111 may calculate a SHAP value (SHapley Additive exPlanations) that evaluates the contribution degree of the explanatory variable, which is a feature quantity, to the model predicted value using SHAP. This SHAP value is used as the risk coefficient a. Regarding SHAP, it is described, for example, in Scott Lundberg, Su-In Lee, “A Unified Approach to Interpreting Model Predictions”, Internet (URL: https: / / arxiv.org / abs / 1705.07874, search date: December 15, 2023).

[0085] However, SHAP is just one example of a method for evaluating the contribution to a model-predicted value, and other methods may also be applied. Methods other than SHAP include, for example, LIME (Local Interpretable Model-agnostic Explanations), Partial Dependence Plots (PDP), and Permutation Importance. A value evaluating the contribution of a feature calculated by these methods may be used as the risk coefficient a.

[0086] Next, the dementia risk analysis processing unit 111 derives risk factors included in the correspondence table 117 that correspond to the input explanatory variable value (step 1213). For example, if a value of "corresponds: 1" is input for the explanatory variable item "BMI<20", the value of that explanatory variable corresponds to the value "BMI<20" of dietary risk factor 803. In this case, the value "1" of risk frequency 804 that corresponds to the value "BMI<20" of dietary risk factor 803 is acquired. In this way, when a value that satisfies the condition set for dietary risk factor 803 is input as an explanatory variable, the value of that dietary risk factor 803 is derived as the risk factor that corresponds to the value of the input explanatory variable.

[0087] Similarly, all values of the dietary risk factor 803, exercise risk factor 903 and social participation risk factor 1003 that correspond to the input explanatory variable values are identified, and the corresponding risk frequency values 804, 904 and 1004 are obtained.

[0088] Then, the dementia risk analysis processing unit 111 calculates a risk intervention frequency based on the calculated risk coefficient and the corresponding acquired risk frequency, and selects an intervention method based on the result (step 1214). The selection result is output to the extraction unit 104.

[0089] Specifically, the dementia risk analysis processing unit 111 calculates the risk intervention frequency for each risk factor by multiplying each risk coefficient a by the risk frequency of the risk factor corresponding to each explanatory variable. Then, depending on whether the risk factor with the highest calculated risk intervention frequency is a dietary risk factor, an exercise risk factor, or a social participation risk factor, the intervention method is selected from dietary intervention, exercise intervention, and social participation intervention. Details of this processing will be described later with reference to FIG. 13.

[0090] The intervention service recommendation analysis processing unit 113 of the extraction unit 104 searches the intervention service provider DB 119 to extract intervention service providers 152 that provide intervention services suitable for the user 140 who is the prediction target, and outputs the extracted information to the user 140. For example, the intervention service recommendation analysis processing unit 113 extracts intervention service providers 152 that provide services for the selected intervention method, and outputs the extracted information to the user 140. At this time, if a target area for the provided service is specified, intervention service providers 152 that provide services targeting an area including the address of the user 140 are extracted. Furthermore, in consideration of user convenience, intervention service providers 152 located in an area close to the address of the user 140 may be preferentially extracted. This solves the problem of it being difficult to understand which hospital or institution to use in one's area when undergoing an examination or intervention, the content of the available services, and the procedures for using them.

[0091] FIG. 13 is an explanatory diagram showing an example of information obtained by the processing executed by the dementia test intervention recommendation system 100 in the embodiment of the present invention.

[0092] 13 shows, as an example, the risk prediction result obtained when the user information, digital biomarker information, and KDB information of the user 140 to be predicted (e.g., the user whose user ID is 0TZfTia5MX) are input as explanatory variables into the dementia risk prediction model. The risk prediction information 1300 includes a risk factor ranking 1301, a risk coefficient 1302, data 1303, a risk factor item 1304, a dietary risk intervention frequency 1305, an exercise risk intervention frequency 1306, and a social participation risk intervention frequency 1307.

[0093] The risk factor item 1304 indicates risk factors that correspond to explanatory variables of the prediction target user 140. Specifically, the risk factor item 1304 lists values of dietary risk factors 803 (FIG. 8), exercise risk factors 903 (FIG. 9), and social participation risk factors 1003 (FIG. 10) that correspond to explanatory variable values of the prediction target user 140.

[0094] Data 1303 indicates the value of data 802 (FIG. 8), data 902 (FIG. 9), or data 1002 (FIG. 10) corresponding to each value of the risk factor item 1304.

[0095] The risk coefficient 1302 indicates the degree of contribution (for example, SHAP value) to the model predicted value calculated for the explanatory variables of the user 140 to be predicted that corresponds to each value of the risk factor item 1304.

[0096] The dietary risk intervention frequency 1305, the exercise risk intervention frequency 1306, and the social participation risk intervention frequency 1307 indicate risk intervention frequencies calculated by multiplying the value of the risk coefficient 1302 by the risk frequency corresponding to each value of the risk factor item 1304. For example, when the value of the risk factor item 1304 is a dietary risk factor, the value obtained by multiplying the value of the risk frequency 804 corresponding to the dietary risk factor by the value of the risk coefficient 1302 is held as the dietary risk intervention frequency 1305. Similarly, when the value of the risk factor item 1304 is an exercise risk factor, the value obtained by multiplying the value of the risk frequency 904 corresponding to the exercise risk factor by the value of the risk coefficient 1302 is held as the exercise risk intervention frequency 1306. Furthermore, when the value of the risk factor item 1304 is a social participation risk factor, the value obtained by multiplying the value of the risk frequency 1004 corresponding to the social participation risk factor by the value of the risk coefficient 1302 is held as the social participation risk intervention frequency 1307.

[0097] The risk factor ranking 1301 indicates the ranking of the magnitude of the risk intervention frequency (i.e., dietary risk intervention frequency 1305, exercise risk intervention frequency 1306, or social participation risk intervention frequency 1307) calculated for each value of the risk factor item 1304. In the risk prediction information 1300 shown in Fig. 13, the values of the risk factor item 1304 are arranged in descending order according to the ranking of the magnitude of the risk intervention frequency.

[0098] Here, an example of the first row of the risk prediction information 1300 shown in Fig. 13 will be described. When the serum albumin level 513 of the user 140 to be predicted is 3.5 g / dl or less and a value indicating this is input as an explanatory variable into the dementia risk prediction model, the dementia risk analysis processing unit 111 calculates a risk coefficient a (e.g., SHAP value) which is the contribution of this explanatory variable to the calculated dementia risk probability value (step 1202). In the example of Fig. 13, the risk coefficient a is calculated to be "0.514".

[0099] Furthermore, the dementia risk analysis processing unit 111 derives that the value of this explanatory variable corresponds to the value of the dietary risk factor 803, "serum albumin level ≦ 3.5 g / dl," and stores this value as the value of the risk factor item 1304. Then, the dementia risk analysis processing unit 111 obtains the value "2" of the risk frequency 804 corresponding to the value "serum albumin level ≦ 3.5 g / dl" of the dietary risk factor 803 (step 1213), and calculates "1.028" as the dietary risk intervention frequency 1305 by multiplying the value "0.514" of the risk coefficient a by the value "2" of the risk frequency 804.

[0100] The dementia risk analysis processing unit 111 performs the same processing as above on the values of other explanatory variables input into the dementia risk prediction model for dementia risk prediction of the prediction target user 140, thereby calculating the dietary risk intervention frequency 1305, exercise risk intervention frequency 1306, or social participation risk intervention frequency 1307 corresponding to each explanatory variable value. In the example of Fig. 13, of the calculated risk intervention frequencies, the value "1.028" of the dietary risk intervention frequency 1305 calculated for "serum albumin level ≦ 3.5 g / dl" is the largest, so the risk factor rank 1301 corresponding to the value of this risk factor item 1304 is set to "1".

[0101] Then, the dementia risk analysis processing unit 111 totals the calculated dietary risk intervention frequency 1305, exercise risk intervention frequency 1306, and social participation risk intervention frequency 1307. In the example of Fig. 13, the total value of the dietary risk factors 803 is "0.626", the total value of the exercise risk intervention frequencies 1306 is "0.067", and the total value of the social participation risk intervention frequencies 1307 is "0.331", and the total value of the dietary risk factors 803 is the maximum.

[0102] As described above, the maximum total value of the dietary risk intervention frequency 1305 for the user 140 being predicted means that if the user 140 develops dementia in the future, the biggest factor is predicted to be a dietary problem (e.g., malnutrition, etc.). In other words, improving dietary problems is predicted to be the most effective way to reduce the dementia risk of the user 140. Therefore, the dementia risk analysis processing unit 111 selects a dietary intervention service as the intervention method (step 1214). The intervention service recommendation analysis processing unit 113 outputs, as an intervention service recommendation, information about an intervention service provider 152 that provides a dietary intervention service to the user 140 (step 1215).

[0103] If the total value of the exercise risk intervention frequency 1306 becomes the maximum as a result of the same processing as above, an exercise intervention service is selected as the intervention method because it is predicted that improving exercise-related problems (e.g., lack of exercise) will be most effective in reducing the dementia risk of the user 140. Similarly, if the total value of the social participation risk intervention frequency 1307 becomes the maximum, it is predicted that improving social participation-related problems (e.g., social isolation) will be most effective in reducing the dementia risk of the user 140, so a social participation intervention service is selected as the intervention method.

[0104] FIG. 14 is an explanatory diagram showing an example of information output by the dementia test intervention recommendation system 100 in an embodiment of the present invention.

[0105] 14 is displayed, for example, on the screen of a smartphone 141 of a user 140 who is the target of prediction. The output information 1400 includes a dementia risk display section 1401, a risk factor display section 1402, an intervention service recommendation display section 1403, an examination recommendation display section 1404, and a detailed examination recommendation display section 1405.

[0106] The dementia risk display section 1401 displays the level of dementia risk (i.e., risk of cognitive decline) calculated for the user 140 who is the prediction target. For example, the dementia risk 1102 value corresponding to the dementia risk probability obtained as a result of inputting the user information and health information of the user 140 who is the prediction target into the dementia risk prediction model as explanatory variables is displayed.

[0107] The risk factor display section 1402 displays, for example, a risk factor predicted as the greatest factor in dementia risk. At this time, the predicted risk factor itself (for example, "serum albumin level ≦ 3.5 g / dl") may be displayed, or information (for example, "nutritional deficiency") that expresses the risk factor in an easy-to-understand manner for the user 140 may be displayed. Similarly, if the predicted risk factor is an exercise risk factor, "lack of exercise" or the like is displayed, and if it is a social participation risk factor, "lack of social participation" or the like is displayed.

[0108] The intervention service recommendation display section 1403 displays information about one or more specific intervention services recommended to the prediction target user 140 from among the intervention services provided by the intervention service provider 152. For example, if a dietary intervention service is selected in step 1214, the names of one or more specific dietary intervention services, a button for displaying details of each, and a button for reserving the use of the intervention service may be displayed.

[0109] Similarly, if an exercise intervention service is selected in step 1214, information about one or more specific exercise intervention services is displayed, and if a social participation intervention service is selected in step 1214, information about one or more specific social participation intervention services is displayed. The user 140 can scan these buttons to display details, select the intervention service they want to receive, and make a reservation. The information displayed at this time is obtained from the intervention service provider DB 119.

[0110] The intervention service recommendation display unit 1403 displays information about one or more specific intervention services that correspond to the recommended intervention service. For example, the names of one or more specific intervention services, buttons for displaying details of each of the services, and buttons for reserving the intervention service may be displayed. The user 140 can operate these buttons to display details, select the intervention service that the user wants to receive, and make a reservation.

[0111] The test recommendation display section 1404 displays information about one or more test services recommended depending on the predicted level of dementia risk. For example, if the dementia risk is predicted to be medium, the names of one or more specific invasive blood test services, buttons for displaying details of each service, and a button for booking the test service may be displayed. The user 140 can operate these buttons to display details, select the test service they wish to receive, and make a booking. The information displayed at this time is obtained from the testing company DB 118 or the local hospital DB 120.

[0112] Information about testing services recommended for more detailed dementia risk assessment is displayed in the detailed test recommendation display section 1405. For example, if the dementia risk is predicted to be medium and an invasive blood test service or the like is recommended as a standard, information about other testing services (such as non-invasive task-based tests and more detailed specialist tests) may be displayed as those recommended for more detailed assessment.

[0113] For example, the name of a task-based testing service, a button for displaying its details, and a button for making a reservation for the testing service may be displayed, or the name of a specialist hospital for performing a specialist test, a button for displaying its details, and a button for making a reservation for the testing service may be displayed. By operating these buttons, the user 140 can display details, select the testing service they want to receive, and make a reservation. The information displayed at this time is obtained from the testing company DB 118 or the local hospital DB 120.

[0114] The above example is aimed at predicting the risk of dementia and recommending tests and interventions based on the results, but by performing similar processing for diseases other than dementia (such as so-called lifestyle-related diseases), it is possible to predict the risk of onset and recommend the next tests and interventions to be performed based on the results.

[0115] Furthermore, the system according to the embodiment of the present invention may be configured as follows.

[0116] (1) A disease testing intervention recommendation method executed by a computer system (e.g., a computer system 200 realizing a dementia testing intervention recommendation system 100) having a processor (e.g., a processor 201) and a storage device (e.g., a memory 202 and an auxiliary storage device 203), wherein the storage device holds health information (e.g., information in a health information DB 116) indicating a person's health condition, and correspondence information (e.g., a correspondence table 117) that associates disease risk factors (e.g., dietary risk factors 803, exercise risk factors 903, or social participation risk factors 1003) with intervention methods (e.g., dietary intervention, exercise intervention, or social participation intervention), and the disease testing intervention recommendation method is carried out by the processor. The method includes a first step (e.g., step 1202) of predicting the disease risk of a person to be predicted by inputting information on multiple items indicating the health condition of the person to be predicted as explanatory variables into a disease risk prediction model that predicts the disease risk of the person based on information on multiple items indicating the condition, and calculating the contribution of each item of the explanatory variables to the predicted disease risk of the person to be predicted (e.g., risk coefficient such as SHAP value); and a second step (e.g., steps 1213 and 1214) in which the processor identifies the disease risk factors corresponding to the input explanatory variables, and selects an intervention method based on the identified disease risk factors, the magnitude of the calculated contribution, and the corresponding information.

[0117] This allows the system to recommend appropriate next intervention methods for the individual depending on the predicted causes of disease risk.

[0118] (2) A method for recommending disease testing interventions as described in (1) above, wherein the correspondence information includes information that associates the disease risk factors with risk frequencies (e.g., risk frequencies of 804, 904, or 1004) that indicate the magnitude of disease risk, and in the second step, the processor calculates a risk intervention frequency for each of the identified disease risk factors by multiplying the risk frequency by the contribution of the item of the explanatory variable corresponding to the disease risk factor, and selects the intervention method based on the calculated risk intervention frequency (e.g., step 1214).

[0119] This allows recommendations for appropriate next steps in intervention depending on the predicted cause of disease risk.

[0120] (3) In the disease testing intervention recommendation method described in (2) above, in the second step, the processor sums the calculated risk intervention frequencies for each corresponding intervention method, and selects the intervention method that maximizes the sum of the risk intervention frequencies.

[0121] This allows recommendations for appropriate next steps in intervention depending on the predicted cause of disease risk.

[0122] (4) The disease testing intervention recommendation method described in (2) above, wherein the storage device further retains intervention service provider information (e.g., information in intervention service provider DB119) for a plurality of intervention service providers, the information including information on the intervention methods each provides, and the disease testing intervention recommendation method further includes a third procedure (e.g., step 1215) in which the processor outputs information on a provider that provides intervention services for the selected intervention method based on the intervention service provider information.

[0123] This will assist in the use of services for the next appropriate intervention depending on the predicted cause of disease risk.

[0124] (5) A disease testing intervention recommendation method as described in (2) above, wherein the disease risk factors include at least one of disease risk factors related to diet (e.g., dietary risk factors 803), disease risk factors related to exercise (e.g., exercise risk factors 903), and disease risk factors related to social participation (e.g., social participation risk factors 1003), and in the correspondence information, the disease risk factors related to diet are associated with dietary intervention (e.g., intervention method 801) as the intervention method, the disease risk factors related to exercise are associated with exercise intervention (e.g., intervention method 901) as the intervention method, and the disease risk factors related to social participation are associated with social participation intervention (e.g., intervention method 1001) as the intervention method.

[0125] This allows recommendations for appropriate next steps in intervention depending on the predicted cause of disease risk.

[0126] (6) The disease testing intervention recommendation method described in (1) above, wherein the correspondence information further includes information that associates the predicted disease risk level with a testing method (e.g., a correspondence table 1100 between dementia risk probability and risk stage), the storage device further stores testing information for multiple testing institutions (e.g., information in testing company DB 118 and information in local hospital DB 120) including information on the testing method each provides, and the disease testing intervention recommendation method further includes a fourth procedure (e.g., steps 1204 to 1207) in which the processor outputs information about the testing institutions that provide tests using the testing method corresponding to the predicted disease risk level.

[0127] This clarifies the predicted level of disease risk for the subject, and makes it possible to recommend an appropriate test method to be performed next depending on the level of the predicted disease risk.

[0128] (7) A disease testing intervention recommendation method as described in (6) above, wherein the testing method includes non-invasive tests (e.g., task-based tests, etc.), invasive tests (e.g., blood tests, etc.), and specialist tests, and non-invasive tests are associated with low disease risks, invasive tests with higher disease risks, and specialist tests with even higher disease risks.

[0129] This allows the system to recommend the appropriate next test method depending on the predicted level of disease risk.

[0130] (8) The method for recommending disease testing interventions described in (1) above further includes a fifth step (e.g., step 1201) in which the processor generates the disease risk prediction model by performing machine learning using one or more disease-related items among the information on multiple items included in the health information as objective variables and multiple other items as explanatory variables.

[0131] This will enable disease risk prediction based on each individual's health information. In particular, it will solve the problem of dementia, which is difficult to distinguish from normal aging and is therefore difficult for the individual and those around them to notice.

[0132] (9) The disease testing intervention recommendation method described in (1) above, wherein the health information includes at least one of digital biomarker information (e.g., digital biomarker information 400) obtained from a device worn by the person, health checkup information (e.g., KDB information 500) obtained by the person undergoing a health checkup, and medical examination information (e.g., KDB information 600) obtained by the person visiting a medical institution.

[0133] This allows disease risk to be predicted using not only health checkup and medical institution consultation information, but also information obtained from devices worn by individuals on a daily basis, making it possible to predict disease risk relatively easily and with high accuracy.

[0134] (10) In the disease test intervention recommendation method described in (1) above, the disease may be dementia.

[0135] This allows the system to recommend the appropriate next intervention method for the individual depending on the predicted cause of dementia risk.

[0136] The present invention is not limited to the above-described embodiments, but includes various modifications. For example, the above-described embodiments have been described in detail to facilitate a better understanding of the present invention, and the present invention is not necessarily limited to those including all of the described configurations. Furthermore, it is possible to replace part of the configuration of one embodiment with the configuration of another embodiment, or to add the configuration of another embodiment to the configuration of one embodiment. Furthermore, it is possible to add, delete, or replace part of the configuration of each embodiment with other configurations.

[0137] Furthermore, the above-described configurations, functions, processing units, processing means, etc. may be partially or entirely implemented in hardware, for example, by designing them as integrated circuits. The above-described configurations, functions, etc. may also be implemented in software, with a processor interpreting and executing a program that implements each function. Information such as the programs, tables, and files that implement each function can be stored in storage devices such as nonvolatile semiconductor memory, hard disk drives, and solid-state drives (SSDs), or in computer-readable, non-transitory data storage media such as IC cards, SD cards, and DVDs.

[0138] In addition, the control lines and information lines shown are those that are considered necessary for the explanation, and not all control lines and information lines in the product are necessarily shown. In reality, it can be considered that almost all components are interconnected. [Explanation of symbols]

[0139] 100 Dementia Testing Intervention Recommendation System 101 Input section 102 Output section 103 Calculation Unit 104 Extraction part 105 Storage section 106 Storage section 130 Communication Network 140 users 150 External Systems

Claims

1. A disease testing intervention recommendation method executed by a computer system having a processor and a storage device, comprising: The storage device includes: health information indicating the health status of the person; Correspondence information that associates disease risk factors with intervention methods is maintained; The disease examination intervention recommendation method includes: a first step in which the processor predicts the disease risk of the person to be predicted by inputting information on a plurality of items indicating the health condition of the person as explanatory variables into a disease risk prediction model that predicts the disease risk of the person based on information on a plurality of items indicating the health condition of the person, and calculates the contribution of each item of the explanatory variables to the prediction result of the disease risk of the person to be predicted; A disease testing intervention recommendation method, comprising: a second step in which the processor identifies the disease risk factors corresponding to the input explanatory variables, and selects an intervention method based on the identified disease risk factors, the calculated magnitude of contribution, and the correspondence information.

2. 2. The disease test intervention recommendation method according to claim 1, the correspondence information includes information that associates the disease risk factors with risk degrees that indicate the magnitude of disease risk, In the second step, the processor calculates a risk intervention frequency for each of the identified disease risk factors by multiplying the contribution of the item of the explanatory variable corresponding to the disease risk factor by the risk frequency, and selects the intervention method based on the calculated risk intervention frequency.

3. 3. The disease test intervention recommendation method according to claim 2, In the second step, the processor sums the calculated risk intervention frequencies for each corresponding intervention method, and selects the intervention method that maximizes the sum of the risk intervention frequencies.

4. 3. The disease test intervention recommendation method according to claim 2, The storage device further stores intervention service provider information including information on an intervention method provided by each of a plurality of intervention service providers; The disease test intervention recommendation method further comprises a third step in which the processor outputs information about a provider that provides intervention services for the selected intervention method based on the intervention service provider information.

5. 3. The disease test intervention recommendation method according to claim 2, the disease risk factors include at least one of diet-related disease risk factors, exercise-related disease risk factors, and social participation-related disease risk factors; A disease testing intervention recommendation method, characterized in that in the correspondence information, dietary intervention is associated as the intervention method for the diet-related disease risk factors, exercise intervention is associated as the intervention method for the exercise-related disease risk factors, and social participation intervention is associated as the intervention method for the social participation disease risk factors.

6. 2. The disease test intervention recommendation method according to claim 1, The correspondence information further includes information that associates the predicted disease risk with a testing method, The storage device further stores test information for a plurality of testing institutions, the test information including information on the test method provided by each of the testing institutions; The disease testing intervention recommendation method further comprises a fourth step in which the processor outputs information about the testing institution that provides testing using the testing method corresponding to the predicted high disease risk.

7. 7. The disease test intervention recommendation method according to claim 6, The testing method includes non-invasive testing, invasive testing, and specialist testing; A disease test intervention recommendation method characterized in that a low disease risk is associated with a non-invasive test, a higher disease risk is associated with an invasive test, and an even higher disease risk is associated with a specialist test.

8. 2. The disease test intervention recommendation method according to claim 1, A disease testing intervention recommendation method characterized by further comprising a fifth step in which the processor generates the disease risk prediction model by performing machine learning using one or more items related to the target disease from among the information on multiple items included in the health information as objective variables and multiple other items as explanatory variables.

9. 2. The disease test intervention recommendation method according to claim 1, A disease testing intervention recommendation method, characterized in that the health information includes at least one of digital biomarker information obtained from a device worn by the person, health checkup information obtained by the person undergoing a health checkup, and medical examination information obtained by the person visiting a medical institution.

10. 2. The disease test intervention recommendation method according to claim 1, The disease examination intervention recommendation method, wherein the disease is dementia.

11. 1. A disease test intervention recommendation device having a processor and a storage device, The storage device health information indicating the health status of the person; Correspondence information that associates disease risk factors with intervention methods is maintained; The processor: A disease risk prediction model predicts a disease risk of a person based on information on a plurality of items indicating the person's health condition, by inputting the information on the plurality of items indicating the person's health condition as explanatory variables, thereby predicting the disease risk of the person to be predicted, and calculating the contribution of each item of the explanatory variables to the prediction result of the disease risk of the person to be predicted; A disease testing intervention recommendation device characterized by identifying the disease risk factors corresponding to the input explanatory variables, and selecting an intervention method based on the identified disease risk factors, the calculated magnitude of contribution, and the correspondence information.

12. A disease testing intervention recommendation program for controlling a computer system having a processor and a storage device, comprising: The storage device health information indicating the health status of the person; Correspondence information that associates disease risk factors with intervention methods is maintained; The disease testing intervention recommendation program a first step of predicting the disease risk of a person to be predicted by inputting information on a plurality of items indicating the health condition of the person as explanatory variables into a disease risk prediction model that predicts the disease risk of the person based on information on a plurality of items indicating the person's health condition, and calculating the contribution of each item of the explanatory variables to the prediction result of the disease risk of the person to be predicted; A disease testing intervention recommendation program characterized by having the processor execute a second step of identifying the disease risk factors corresponding to the input explanatory variables, and selecting an intervention method based on the identified disease risk factors, the calculated magnitude of contribution, and the correspondence information.

Images