Information Processing Apparatus, Information Processing Method, and Program

Abstract

It aims to improve the accuracy and convenience of providing health-related information. The model acquisition unit 51 is used to acquire a model that outputs an index related to the lifespan of the target person when the health-related information of the target person is input as the all-cause mortality model 4, and this model is obtained by learning using the health-related information of each of multiple people belonging to the first country and the first statistical information on the lifespan of the first country. The statistical information acquisition unit 52 is used to acquire the first statistical information of the second country to which the target person belongs. The health-related information acquisition unit 53 is used to acquire the health-related information of the target person. The first type of score calculation unit 54 calculates the first type of score related to the lifespan of the target person based on the index output as a result of inputting the acquired health-related information into the all-cause mortality model 4 and the acquired first statistical information.

Description

Technical Field

[0001] The present invention relates to an information processing apparatus, an information processing method, and a program. Background Art

[0002] Conventionally, there has been a technique for presenting the risk of a predetermined disease to a user (for example, refer to Patent Document 1).

[0003] Prior Art Documents

[0004] Patent Documents

[0005] Patent Document 1: Japanese Patent Application Laid-Open No. 2018-005726 Summary of the Invention

[0006] Problems to be Solved by the Invention

[0007] However, in the prior art including the technique described in Patent Document 1, only risk factors for developing the predetermined disease are extracted based on information of patients with the predetermined disease as analysis targets, and users having such risk factors are informed of a high risk.

[0008] It is desired to present various types of health-related results (information). That is, it is desired to provide comprehensive health-related information. In addition, for example, in recent years, it has also been desired to improve the accuracy of health-related information.

[0009] An object of the present invention is to improve the accuracy and convenience of providing health-related information.

[0010] Means for Solving the Problems

[0011] To achieve the above object, an information processing apparatus according to one aspect of the present invention includes:

[0012] a model acquisition unit that acquires, as an all-cause mortality model, a model that outputs an index related to the life expectancy of an object person when health-related information of the object person is input, the model being obtained by learning using the health-related information of each of a plurality of persons belonging to a first population and first statistical information on the life expectancy of the first population;

[0013] a statistical information acquisition unit that acquires the first statistical information of a second population to which the object person belongs;

[0014] a health-related information acquisition unit that acquires the health-related information of the object person;

[0015] a first type score calculation unit that calculates a first type score related to the life expectancy of the object person based on the index output as a result of inputting the acquired health-related information into the all-cause mortality model and the acquired first statistical information.

[0016] The information processing method and program according to one aspect of the present invention respectively correspond to the information processing method and program of the information processing apparatus according to one aspect of the present invention described above.

[0017] Effects of the Invention

[0018] According to the present invention, it is possible to improve the accuracy and convenience of providing health-related information. Description of the Drawings

[0019] Figure 1 It is a schematic diagram showing an example of the outline of the service of an information processing system including a determination device to which one embodiment of the present invention is applied.

[0020] Figure 2 It is a diagram showing a configuration example of an information processing system including a determination device according to one embodiment of the present invention.

[0021] Figure 3 It shows Figure 2 An example of a block diagram of the hardware configuration of the determination device of one embodiment of the information processing device of the present invention in the information processing system.

[0022] Figure 4 It shows Figure 3 An example of a functional block diagram of the functional structure of the determination device.

[0023] Figure 5 It is for explaining an example of a flowchart of a request estimation process executed by a server having Figure 4 the functional structure.

[0024] Figure 6 It shows Figure 1 A conceptual diagram of the learning process in the learning device.

[0025] Figure 7 It conceptually shows a diagram of the type of score provided by a determination device having Figure 4 the functional structure. Detailed Description of the Invention

[0026] Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[0027] Figure 1 It is a schematic diagram showing an example of the outline of the service (hereinafter referred to as "this service") of an information processing system including a determination device to which one embodiment of the present invention is applied.

[0028] This service is a service that provides information related to the health status of a user.

[0029] Specifically, this service is provided by a service provider (not shown) and used by a user and a user who provides information related to the user's health status with the user's permission (e.g., an insurance company, an employer, etc. that the user plans to contract with). The following description assumes that the user is the user of this service.

[0030] Here, the service provider (not shown) provides this service to the user using the user terminal 3 by managing the learning device 1 and the determination device 2.

[0031] The following will Figure 1 describe the outline of the process of this service along steps ST1 to ST7.

[0032] In step ST1, the learning device 1 performs preprocessing.

[0033] Here, the learning device 1 uses a predetermined algorithm to perform a learning process in order to generate or update the all-cause mortality model 4 and the individual model 5 described later.

[0034] It should be noted that although the description assumes that a heuristic model construction algorithm is used as the predetermined algorithm, any algorithm can be used as the predetermined algorithm, such as an algorithm using machine learning or an algorithm for an integrated model, etc.

[0035] The learning device 1 uses the following data in the learning process.

[0036] That is, the learning device 1 uses the "health-related parameters" of each person among multiple people belonging to the first country to perform the learning process.

[0037] The "health-related parameters" of a certain person are parameters considered to be related to the health of that person. Specifically, for example, the "health-related parameters" may include items such as basic background information (including gender and age), annual health examination data, core risk factors, lifestyle data, family history, etc. However, the health-related parameters are not limited to these. In particular, in this service, if new health-related parameters are found or deleted in evidence such as papers, any item can be flexibly added or reduced.

[0038] In addition, depending on the person as the object, the health-related parameters may be information on all items or some items. In this service, information on the accuracy of the determination process result using the learning result can also be presented to the user according to the items of the health-related parameters.

[0039] In addition, the learning device 1 uses the "statistical data" of the first country to perform the learning process.

[0040] That is, the "statistical data" is statistical data related to the health of a population corresponding to multiple people within a predetermined range.

[0041] Specifically, for example, the statistical data may include items such as lifespan statistical data (life tables), mortality rankings, distributions (bandwidths), and boundary analysis of predicted values. However, the statistical data is not limited to these. As will be described in detail later, in this service, lifespan statistical data can be used to present a lifespan-related score to the user, but scores related to other health conditions can also be presented to the user. The statistical data can also include various types of statistical data used to calculate scores for other health conditions.

[0042] In addition, boundary analysis refers to a new method of incorporating the uncertainty of parameters without assuming a specific form of the distribution function. The all-cause mortality model and the method using statistical data described later are specific examples of boundary analysis.

[0043] Here, the learning device 1 is characterized by using the statistical data of the first country. That is, the learning device 1 can perform a learning process that closely matches the people belonging to the first country by using the "health-related parameters" of each of multiple people belonging to the first country and the statistical data of the first country. Thereby, a model capable of performing an inference process that closely matches the people belonging to the first country is generated or updated.

[0044] It should be noted that the learning device 1 also uses "various health data" in the learning process. That is, the statistical data used by the learning device 1 includes not only the statistical data of a specific country such as the first country as various health data. In addition, for example, in the learning process, the learning device 1 uses papers published around the world and information related to the influence of exercise volume, alcohol consumption, diseases, and test values on lifespan as various health-related data. That is, the learning device 1 uses statistical data not limited to the first country as evidence in model generation to achieve a more accurate learning process.

[0045] In preprocessing, the learning device 1 performs data mapping, data alignment, data cleaning, and partitioning into respective data sets.

[0046] Here, data mapping is a process of adjusting the original data according to the learning process. Specifically, for example, it refers to a process of matching data standards to achieve integration between databases.

[0047] In addition, data alignment refers to a process of configuring the data within the device in a design-compliant manner for the purpose of improving data access efficiency.

[0048] In addition, data cleaning refers to noise data removal and name matching processing.

[0049] In addition, partitioning into respective data sets refers to a process of partitioning into those for learning, testing, verification checks, etc.

[0050] In this way, the health-related parameters and statistical data are processed through preprocessing, so as to effectively perform the learning process.

[0051] Next, in step ST2, the learning device 1 generates a model based on the training data and test data in the preprocessed health-related parameters, the statistical data of the first country, and various statistical data.

[0052] Specifically, the learning device 1 generates the all-cause death model 4 of the people belonging to the first country by using at least the health-related parameters of each of multiple people belonging to the first country and the statistical data of the first country.

[0053] The all-cause death model refers to a model that outputs information indicating the risk position of the lifespan of the target person in the lifespan distribution of the population based on the health-related parameters of the person to be processed.

[0054] Specifically, for example, according to the all-cause death model, when inputting health-related parameters indicating the health of the target person, output position information such as the top 10% in the lifespan distribution of the population in the first country. It should be noted that the position information can be in any form. Specifically, for example, the position information can be ranking information such as the 10th among 100 people.

[0055] In the heuristic model construction algorithm, "make judgments based on rules of thumb and preconceived notions". That is, a model is generated based on whether data consistent with the generally recognized trends based on papers, etc. can be obtained. For example, by confirming that there are no contradictions among viewpoints such as "although in the same health condition, women have a shorter lifespan than men", "although the average lifespan gradually increases year by year, the future life expectancy of the young is expected to be shorter than that of the elderly", "a large amount of alcohol consumption has a positive effect on prolonging lifespan", "information such as the degree of abnormality of test values and the degree of diseases reported in clinical reports", etc. to generate a model.

[0056] Next, in step ST3, the learning device 1 performs a verification check on the generated all-cause death model 4 based on the verification data in the preprocessed health-related parameters. Thereby, the accuracy of the all-cause death model 4 and the certainty of the result (for example, the average error) can be calculated.

[0057] For example, verification is performed by comparing the actual lifespan of the participant data (deceased people from the first country) with known lifespans with the predicted values calculated according to the model.

[0058] Next, in step ST4, the learning device 1 updates the all-cause death model 4. That is, when the data volume of the health-related parameters and statistical data increases or the model is updated based on new papers, etc., the all-cause death model 4 is updated and its accuracy is improved.

[0059] Although not shown, an individual model 5 is generated or updated for a predetermined "health condition" by a process similar to the above-described steps ST1 to ST4.

[0060] Here, the "health condition" refers to a predetermined "disease", "conceptual interpretation", "composite concept", and "other health outcome-related information".

[0061] Here, the "conceptual interpretation" and "composite concept" refer to scores that vividly express health conditions, health awareness, etc. Specifically, for example, they include "heart age", "skin age", "lifestyle score", etc. That is, the heart age represents the risk of a heart attack or stroke. In addition, the skin age is a value representing the age level of the skin. In addition, the lifestyle score is the sum of scores for alcohol consumption, smoking, exercise, etc. Thus, the conceptual interpretation is a score interpreted according to a predetermined concept. In addition, the composite concept is a score that combines multiple factors.

[0062] That is, as will be described in detail later, in this service, information related to the risk of developing a predetermined disease is presented to the user. In addition, various health outcome-related information, such as conceptual interpretations and composite concepts, is presented to the user.

[0063] Next, in step ST5, the determination device 2 performs a determination process related to lifespan on the health scores presented to the user who is the processing target.

[0064] Here, the "health score" quantifies health to improve long-term health outcomes and predict short-term and long-term mortality rates as well as health condition risks. Specific examples of health scores will be described later.

[0065] In this service, a user who is the processing target can provide a health score to a user belonging to a second country, which is different from the first country where the learning process was performed. Here, it is normal for there to be statistical differences in data such as the medical provision situation and the basic average lifespan between the first country and the second country.

[0066] The determination device 2 uses the following data in the determination process.

[0067] The determination device 2 performs the determination process using the "health-related parameters" of the user belonging to the second country.

[0068] In addition, the determination device 2 performs the determination process using the "statistical data" of the second country.

[0069] Thus, in the determination process, health-related information of the user belonging to the second country and the statistical data of the second country are used. Thus, the all-cause mortality model 4 generated or updated by using the health-related information of people belonging to the first country and the statistical data of the first country through the learning process can be applied to users in the second country.

[0070] Through life determination, the main part of the health score presented to the user, that is, the information related to life, is generated.

[0071] In addition, in step ST6, the score related to the predetermined health condition of the user is determined based on the individual model 5, which is basically similar to the above step ST5.

[0072] Through health condition determination, the secondary part of the health score presented to the user, that is, the risks (absolute, categorical, relative) related to predetermined diseases, etc., the frequency of verbal explanation, the concept explanation, and the information related to the composite concept are generated.

[0073] Next, in step ST7, the determination device 2 generates information about the improvement actions of the user.

[0074] That is, based on the user's life span, the calculation results of various scores, the calculation process information, etc., the information about the actions for improving the factors (such as disease risks) that affect the user's life span is generated.

[0075] Then, the generated improvement action information is presented to the user.

[0076] In this way, the health score and the improvement action information for improving the health score are presented to the user terminal 3 of the user belonging to the second country. The health score mainly includes the information related to life span and includes the predetermined health condition information.

[0077] Then, in this service, the information of people (users) from different countries can be used for processing in the learning process and the determination process. Thus, the following effects are obtained.

[0078] That is, in the learning process, the accuracy of the model is improved by having a large amount of reliable health-related parameters and statistical data. Therefore, as the first country, the data of a country that can conduct good medical-related paper research can be used.

[0079] However, as described above, even if the determination process is performed using such a generated model and the health-related parameters of users from other countries (the second country), accurate determination results cannot be obtained. Therefore, in this service, by using the statistical data of the first country in the learning process and the statistical data of the second country in the determination process, the users of the second country can be determined using the model of the first country.

[0080] Here, when using the statistical data of the first country in the learning process and the statistical data of the second country in the determination process, the all-cause mortality model 4 and the individual model 5 are models that are indicators of the positions within the distribution of the output statistical data, respectively. That is, based on the health status parameters, the position among the people (population) in that country is calculated. Then, based on this position, in order to reproduce the distribution of the second country, it is determined what specific score (life expectancy) the user at this position has in the distribution.

[0081] Figure 2 FIG. is a diagram showing a configuration example of an information processing system including a determination device according to an embodiment of the present invention.

[0082] Figure 2 The information processing system shown is configured to include a learning device 1, a determination device 2, a user terminal 3, an all-cause mortality model 4, and an individual model 5.

[0083] The learning device 1 generates or updates the all-cause mortality model 4 and the individual model 5 by executing the processing of steps ST1 to ST4 of Figure 1 .

[0084] The determination device 2 obtains the (or updated) models obtained as the learning results of the learning device 1 from the all-cause mortality model 4 and the individual model 5, and determines the life expectancy and health status of the user. Details of the functional structure and processing of the determination device 2 will be described later with reference to Figure 4 etc.

[0085] The user terminal 3 is an information processing device used by the user, which receives user input of health-related information or allows an operation to provide it to the determination device 2.

[0086] The all-cause mortality model 4 stores model data, which is a model that outputs information indicating the risk position of the life expectancy of the target person in the population life expectancy distribution based on the health-related parameters of the person to be processed.

[0087] The individual model 5 stores model data, which is a model that outputs health outcome-related information such as a predetermined disease, a concept explanation, and a composite concept.

[0088] Figure 3 FIG. is a block diagram showing an example of the hardware configuration of a determination device according to an embodiment of the information processing device of the present invention in the Figure 2 information processing system.

[0089] The determination device 2 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, a bus 14, an input / output interface 15, an output unit 16, an input unit 17, a storage unit 18, a communication unit 19, and a driver 20.

[0090] The CPU 11 executes various processes according to a program recorded in the ROM 12 or a program loaded from the storage unit 18 into the RAM 13.

[0091] The RAM 13 also appropriately stores data and the like required for the CPU 11 to execute various processes.

[0092] The CPU 11, the ROM 12, and the RAM 13 are interconnected via the bus 14. The input / output interface 15 is also connected to the bus 14. The input unit 16, the output unit 17, the storage unit 18, the communication unit 19, and the driver 20 are connected to the input / output interface 15.

[0093] The input unit 16 is composed of a keyboard, a mouse, etc., and inputs various information according to the operations of a user's instructions.

[0094] The output unit 17 is composed of a display, a speaker, etc., and outputs images and audio.

[0095] The storage unit 18 is composed of a hard disk, etc., and stores data of various information.

[0096] The communication unit 19 controls communication with other terminals (e.g., Figure 1 the model DB 2 in

[0097] A removable medium 31 such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory is appropriately installed in the driver 20. A program read by the driver 20 from the removable medium 31 is installed in the storage unit 18 as needed. In addition, the removable medium 31 can also store various data stored in the storage unit 18 in the same manner as the storage unit 18.

[0098] It should be noted that, although not shown, Figure 2 the learning device 1 of the information processing system in Figure 4 has substantially the same hardware configuration as that shown in

[0099] In addition, Figure 2 the user terminal 3 of the information processing system in Figure 4 has substantially the same hardware configuration as that shown in

[0100] In addition, for ease of explanation, the learning device 1 and the determination device 2 are provided separately, but the present invention is not limited thereto, and each function of the learning device 1 and the determination device 2 may be integrated into one information processing device.

[0101] In addition, in Figure 2 the all-cause mortality model 4 and the individual model 5 are provided separately from the learning device 1 and the determination device 2, but may be provided, for example, in a part of the storage unit 18 of the determination device 2.

[0102] Figure 4 is a functional block diagram showing an example of the functional structure of the determination device Figure 3 of

[0103] The model acquisition unit 51 acquires information on the all-cause mortality model 4 that outputs an index related to the life expectancy of the target person when the health-related information of the target person is input. This model is obtained by learning using the health-related information of each person among a plurality of people belonging to the first country and the first statistical information on the life expectancy of at least the first country. It should be noted that, as described above, statistical data other than that of the first country is also appropriately used in the learning process of the all-cause mortality model 4.

[0104] In addition, the model acquisition unit 51 also acquires the individual model 5 that outputs an index related to the predetermined health condition of the target person when the health-related information of the target person is input. This model is obtained by learning using the health-related information of each person among a plurality of people belonging to the first country and the second statistical information on the predetermined health condition of the first country.

[0105] Here, the index is the position of the target person in the population. Specifically, for example, the index related to life expectancy may be the position of the target person in the estimated life expectancy distribution of the population, corresponding to the ranking of the death risk.

[0106] The statistical information acquisition unit 52 acquires the first statistical information related to the life expectancy of the second country to which the target person belongs. Specifically, for example, the first statistical information includes information on the life table regarding life expectancy. Here, the life table is a table in demography that shows the mortality rate and average life expectancy of a specific age group and gender.

[0107] In addition, the statistical information acquisition unit 52 also acquires the second statistical information related to the predetermined health condition of the second country to which the target person belongs.

[0108] It should be noted that the statistical information is not limited to the life table, and also includes life expectancy, prognosis, disease distribution, incidence rate, number of patients, etc. from government announcements and published papers.

[0109] The health-related information acquisition unit 53 acquires health-related information of the user. Here, the health-related information includes at least one of background information, health examination information, risk factor information, lifestyle information, family history information, current health status, medication status, and medical history.

[0110] The first type of score calculation unit 54 calculates, for the user as the subject person, a first type of score related to the lifespan of the user based on an index output as a result of inputting the acquired health-related information into the all-cause mortality model 4 and the first statistical information of the acquired second country.

[0111] The first type of score calculation unit 54 calculates, as the first type of score, a score related to the lifespan of the subject person based on the life table of the second country and the position of the subject person in the second country.

[0112] Here, the first type of score may include the lifespan itself or the EBHS (EVIDENCE - BASED HEALTH SCORE). The EBHS score is a value obtained by dividing the predicted remaining lifespan of the subject person by the average lifespan of people of the same age as the subject person in the second country.

[0113] That is, the first type of score calculation unit 54 predicts the remaining lifespan of the subject person based on the life table of the second country and the position of the subject person in the second country, and calculates, as the EBHS score, a value obtained by dividing the predicted remaining lifespan of the subject person by the average lifespan of people of the same age as the subject person in the second country.

[0114] An EBHS score of 100 means being at the average level, greater than 100 means having a longer lifespan than the average, and less than 100 means having a shorter lifespan than the average.

[0115] The second type of score calculation unit 55 calculates a second type of score related to the predetermined health status of the subject person based on an index output as a result of inputting the health-related information of the subject person acquired by the health-related information acquisition unit 53 into the individual model 5 and the acquired second statistical information.

[0116] Here, the predetermined health status is one or more health statuses among diseases, concept explanations, and composite concepts. That is, the second type of score includes scores related to each of the one or more health statuses.

[0117] The improvement activity generation unit 56 generates information related to improvement activities based on at least a part of the output results of the all-cause mortality model 4 and the individual model 5, and the improvement activities indicate the activities that the subject person should perform to improve the first type of score.

[0118] The information providing control unit 57 performs control to provide the information including the first type of score to the user terminal 3.

[0119] The information providing control unit 57 may also perform control to provide information related to the second type of score and improvement actions to the user terminal 2.

[0120] Figure 5 is a diagram for explaining the outline of the input / output in a determination device having a Figure 4 functional structure.

[0121] As Figure 5 shown, the input includes health-related parameters. The health-related parameters include 2 demographic items, 47 annual health check results, 1 other risk factor, 3 lifestyle risk factors, and 7 family histories. Then, by adding parameters that contribute to improving future accuracy as additional factors, the input can be flexibly processed.

[0122] Based on this input, a detailed data method is used in this service to obtain a highly accurate output. That is, a heuristic method is used as a search component for the impact on the health score. That is, a model is generated according to whether data consistent with the generally recognized trend based on papers, etc. can be obtained.

[0123] Thus, by using the above health-related parameters consisting of a large number of items as input, an effective all-cause mortality model 4 and individual model 5 can be constructed using the heuristic method of this input.

[0124] Then, the all-cause mortality model 4 and individual model 5 generated through the learning process in this way are used to generate an output.

[0125] Here, the all-cause mortality model 4 and individual model 5 are calculated based on data of over 100 million person-years, including the statistical data of the first country and various statistical data. That is, as described above, since the learning process is performed based on the data of the first country where sufficient data has been accumulated, the accuracy of the learning models (all-cause mortality model 4 and individual model 5) is improved.

[0126] It should be noted that although the description is omitted here, as used above Figure 1 and Figure 4 described, in the determination process of this service, models can be appropriately used to make determinations for users in the second country. That is, even for users in countries where sufficient data has not been accumulated, this service can make appropriate determinations.

[0127] Then, in the output, the score related to lifespan is provided to the user as the main item. Specifically, for example, the lifespan itself (e.g., Figure 5 90 years old in the example of

[0128] In addition, in the output, the sub-scores are configurable. That is, the user can make the user terminal 3 display information related to predetermined diseases, concept explanations, composite concepts, and other health outcomes that the user himself expects. In addition, the display format can also be freely set.

[0129] Thus, based on lifespan, the user can participate in activities to extend lifespan by mastering information related to diseases, concept explanations, composite concepts, and other health outcomes that may affect that lifespan.

[0130] Figure 6 Is a Figure 1 conceptual diagram showing the learning process in the learning device.

[0131] As Figure 6 shown, first, health parameters of multiple people in a first country are obtained as learning data. Then, model construction (generation or update) is performed. At this time, mapping, alignment, data cleaning, and partitioning into respective data sets are performed to prepare the raw data. Thus, as described above, the all-cause mortality model 4 is generated.

[0132] Next, the all-cause mortality model 4 is used to perform a heuristic analysis of lifespan. At this time, information on mortality rankings, life tables, distribution bandwidths, and prediction value boundaries is used for heuristic adjustment and verification. Thus, a heuristic analysis of lifespan is performed according to the all-cause mortality model 4, and thus lifespan can be calculated.

[0133] Next, a verification check is performed. That is, it is verified whether the verification results of the all-cause mortality model 4 and the heuristic lifespan are valid. If it is determined through verification that the accuracy is low, the model is reconstructed and iteratively improved. In addition, if it is determined that the accuracy is high, the mortality model 4 is adopted. However, even if a high-accuracy all-cause mortality model 4 is obtained, as the learning data is updated, the user data increases, etc., the model will be appropriately reconstructed (updated, iteratively improved).

[0134] Figure 7 Is a diagram conceptually showing the types of scores provided by the determination device having the Figure 4 functional structure.

[0135] As Figure 7 shown, the types of scores can include "lifespan", "absolute risk", "categorical risk", "relative risk", "frequency of verbal explanation", "concept explanation and composite concept".

[0136] Here, lifespan is the most important (key) score.

[0137] That is, lifespan reflects various diseases and health conditions, and improving lifespan or the above-mentioned EBHS score is the goal that the user himself should pursue.

[0138] Then, as information provided in addition to lifespan, "absolute risk", "categorical risk", and "relative risk" are the risks of developing a predetermined disease or the like. Risks can have multiple forms, such as absolute, categorical, and relative.

[0139] The absolute, categorical, and relative forms can be selected by the user or can be changed according to the type of disease and the accuracy of the determination.

[0140] In addition, explaining frequency verbally means explaining the frequency of a predetermined action in sentences. Specifically, for example, explanations such as "drink a large amount of alcohol", "should halve the amount of alcohol consumed", "sufficient exercise", and "should walk for 1 hour every day" are all examples of explaining frequency verbally. In addition, for example, simple words praising the amount of exercise, such as "well done", are also examples of explaining frequency verbally. It should be noted that not only character strings can be used, but also emojis or icons corresponding to the characters can be used.

[0141] Above, an embodiment of the present invention has been described, but the present invention is not limited to the above embodiment, and any modifications, improvements, etc. are included in the present invention as long as the object of the present invention can be achieved.

[0142] For example, the improvement action generation unit 56 can present improvement actions as follows.

[0143] That is, for example, first, the improvement action generation unit 56 calculates the influence value of each factor of the health-related information on the first type of score (lifespan itself or EBHS) for each target person.

[0144] Here, as described above, the health-related information includes factors, which are variables used when calculating the first type of score (lifespan itself or EBHS), including background information, health examination information, risk factor information, lifestyle information, family history information, etc.

[0145] The improvement action generation unit 56 can use any method to calculate the influence value of each factor of the health-related information on the first type of score (lifespan itself or EBHS) for each target person.

[0146] Specifically, for example, as a first method, the improvement action generation unit 56 can calculate the influence value by focusing on samples other than the target person himself, extracting the K nearest neighbors (kNearest Neighbor) close to the target person himself and comparing them with the whole.

[0147] That is, the first method is to select a group for each factor whose factor value is only close to oneself, and consider that the group whose lifespan is significantly different from the whole compared to oneself is an important factor for oneself.

[0148] Alternatively, for example, as a second method, the improvement action generation unit 56 can calculate the degree of influence of the value of the factor by focusing on the all-cause mortality model 4 itself and observing the change in the first-type score when the value of the factor is slightly changed.

[0149] That is, the second method is to focus on the model itself for calculating life expectancy and consider that the factors that cause large fluctuations in life expectancy when only the value of the factor is changed are important factors.

[0150] In the second method, a predetermined LIME (Local Interpretable Model-agnostic Explanations) can be used.

[0151] Alternatively, for example, as a third method, during the learning process of the all-cause mortality model 4, the improvement action generation unit 56 can calculate the degree of influence of the value of the factor based on how much the learning accuracy of the model is improved due to the characteristics of the factor.

[0152] That is, in the third method, the overall importance of the all-cause mortality model 4 is defined, and if the factor has no value, the performance fluctuation during the learning process is large, and the factors with higher importance during the learning process can be calculated as having a greater impact on the first-type score due to the value of the factor.

[0153] In the third method, permutation importance, etc. can be used.

[0154] It should be noted that as described above, the improvement action generation unit 56 can use any method to calculate the influence value of each factor of the health-related information on the first-type score (life expectancy itself or EBHS) for each target person. In addition, the improvement action generation unit 56 can also calculate the influence value of each factor of the health-related information on the first-type score (life expectancy itself or EBHS) based on the results of multiple methods.

[0155] In this way, the improvement action generation unit 56 calculates the influence value of each factor of the health-related information on the first-type score (life expectancy itself or EBHS) for each target person.

[0156] Then, the improvement action generation unit 56 can present the influence value and the information generated based on the influence value to the target person as information related to the improvement action.

[0157] Specifically, for example, for Mr. A, in addition to the first-type score, that is, EBHS is 107.8 and life expectancy is 90.82 years old, the influence value of exercise habit can also be presented as 3.5, the influence value of systolic blood pressure can be presented as 4.2, the influence value of fasting blood glucose can be presented as 2.7, etc. for each factor as information related to the improvement action.

[0158] Thus, Mr. A can understand that he can improve the first type of score by taking improvement actions on the factors that have a higher impact value on his first type of score.

[0159] In addition, in this service, the information of the first type of score, the second type of score, and the improvement actions can be presented to the administrator of the organization to which the subject person belongs (for example, the person in charge of the human resources department, the industrial doctor, etc.). That is, the administrator can grasp the information of the first type of score, the second type of score, and the improvement actions of each subject person belonging to the organization.

[0160] Specifically, for example, statistical information comparing the population of subject persons belonging to the organization with other populations (for example, the population of their own country) is presented to the administrator of the organization.

[0161] In addition, for example, the information of the first type of score, the second type of score, and the improvement actions of each person among the subject persons belonging to the organization is presented to the administrator of the organization.

[0162] Thus, the administrator of the organization can use it to manage the subject persons within the organization, thereby creating a better environment for the entire organization.

[0163] In the description of the above embodiment, the first country in the life determination learning process and the second country in the inference process are the same as the first country in the life determination learning process and the second country in the inference process, but it is not particularly limited to this.

[0164] That is, for example, the statistical data of (at least) the 1-1 country can be used in the learning process of the all-death model 4, and the statistical data of the 1-2 country different from (at least) the 1-1 country can be used in the learning process of the individual model 5.

[0165] In addition, in the description of the above embodiment, the individual model 5 is generated or updated through the learning process, but it is not particularly limited to this. The individual model 5 can adopt a predetermined model generated or updated by a system other than the information processing system (learning device 1 or determination device 2) of the present invention. That is, in the determination process, in addition to the individual model 5 related to a certain health condition generated or updated through the learning process, an individual model 5 related to another health condition generated or updated by another system can also be adopted. Thus, the subject person who is a user of this service can confirm the inference results of the individual model 5 related to various health conditions as health scores.

[0166] In addition, in the description of the above-described embodiment, the score and improvement actions are presented to the user through the user terminal 3, but it is not particularly limited thereto. That is, for example, the score and improvement actions may be recorded on a predetermined medium and presented to the user. Specifically, for example, the service provider may present the score and improvement actions determined by the determination device 2 by printing them on a paper medium and providing it to the user.

[0167] For example, the above-described series of processes may be executed by hardware or software.

[0168] In other words, Figure 4 the functional structure in [] is merely an example and is not particularly limited.

[0169] That is, the information processing system only needs to have a function capable of executing the above-described series of processes as a whole, and it is not particularly limited to Figure 4 the example shown. In addition, the location of the functional block is not specifically limited to Figure 4 the location shown, and it can be arbitrary. For example, the functional block of the determination device 2 may be transferred to the learning device 1 or the like. In addition, the functional block of the learning device 1 may be transferred to the determination device 2 or the like. Furthermore, the learning device 1 and the determination device 2 may be the same hardware.

[0170] In addition, for example, when a series of processes are executed by software, the program constituting the software is installed on a computer or the like from a network or a recording medium.

[0171] The computer may be a computer built into dedicated hardware.

[0172] In addition, the computer may be a computer capable of executing various functions by installing various programs, such as a server, a general-purpose smart phone, or a personal computer.

[0173] In addition, for example, the recording medium containing such a program is not only a removable medium (not shown) distributed separately from the device main body to provide the program to the user, but also a recording medium or the like provided to the user in a state pre-installed in the device main body.

[0174] It should be noted that in this specification, the steps of writing a program to be recorded on a recording medium include not only processes executed in chronological order, but also processes that are not necessarily executed in chronological order but are executed in parallel or individually.

[0175] In addition, in this specification, the term "system" refers to an overall device composed of a plurality of devices, a plurality of units, etc.

[0176] In other words, the information processing device applying the present invention can adopt various embodiments having the following configurations.

[0177] It suffices to include the following units:

[0178] A model acquisition unit (e.g., the model acquisition section 51 in Figure 4 ) for acquiring, as an all-cause mortality model, a model that outputs an index related to the lifespan of an object person when health-related information of the object person is input (e.g., the all-cause mortality model 4 in Figure 4 ), which is obtained by learning using the health-related information of each person among multiple people belonging to a first population (e.g., a first country) and first statistical information on the lifespan of the first population;

[0179] A statistical information acquisition unit (e.g., the statistical information acquisition section 52) for acquiring the first statistical information of a second population (e.g., a second country) to which the object person belongs;

[0180] A health-related information acquisition unit (e.g., the health-related information acquisition section 53) for acquiring the health-related information of the object person;

[0181] A first type of score calculation unit (e.g., the first type of score calculation section 54) for calculating a first type of score related to the lifespan of the object person (e.g., lifespan) based on the index output as a result of inputting the acquired health-related information into the all-cause mortality model and the acquired first statistical information.

[0182] Thus, a model can be created based on the health-related information of people belonging to a first country and the statistical information related to the first country, and an object person belonging to a second country can be accurately determined.

[0183] That is, for example, by learning based on the data of a first country with highly statistical health-related information, an object person in a second country with statistical data different from those of the first country can be accurately determined.

[0184] The health-related information may include at least one of background information, health examination information, risk factor information, lifestyle information, family history information, current health status, medication status, and medical history.

[0185] The first statistical information may at least include information on the life table of lifespan.

[0186] The index related to the lifespan is the position of the object person in the population estimated lifespan distribution,

[0187] The first type of score calculation unit may calculate a score related to the lifespan of the object person as the first type of score based on the life table of the second population and the position of the object person in the second population.

[0188] The first type of score calculation unit can predict the remaining lifespan of the target person based on the life table of the second population and the position of the target person in the second population, and calculate a value obtained by dividing the predicted remaining lifespan of the target person by the average lifespan of people of the same age as the target person in the second population as the first type of score.

[0189] It may further include an information providing control unit (e.g., Figure 4 the information providing control section 57 in

[0190] ), which performs control to provide information including the first type of score to the terminal of the target person.

[0191] The model acquisition unit further acquires, as a health status model, a model that outputs an index related to the predetermined health status of the target person when the health-related information of the target person is input, and the model is obtained by learning using the health-related information of each of the multiple people belonging to the first population and second statistical information on the predetermined health status of the first population;

[0192] The statistical information acquisition unit further acquires the second statistical information of the second population to which the target person belongs;

[0193] The second type of score calculation unit calculates a second type of score related to the predetermined health status of the target person based on the index output as a result of inputting the health-related information of the target person acquired by the health-related information acquisition unit into the health status model and the acquired second statistical information,

[0194] The information providing control unit may also perform control to provide the information including the second type of score to the terminal of the target person.

[0195] The predetermined health status is one or more health statuses among diseases, concept explanations, and composite concepts,

[0196] The second type of score may include scores related to each of the one or more health statuses.

[0197] It further includes an improvement activity information generation unit (e.g., Figure 4 the improvement activity generation section 56 in

[0198] The information providing control unit may also perform control to provide the information including information related to the improvement activity to the terminal of the target person.

[0199] Explanation of reference numerals

[0200] 1... Learning device, 2... Determination device, 3... User terminal, 4... All-death model, 5... Individual model, 6... Statistical information DB, 11... CPU, 51... Model acquisition unit, 52... Statistical information acquisition unit, 53... Health-related information acquisition unit, 54... First type score calculation unit, 55... Second type score calculation unit, 56... Improvement activity generation unit, 57... Information providing control.

Claims

1. An information processing apparatus, wherein, Comprising: a model acquisition unit configured to acquire, as an all-cause mortality model, a model that outputs an index related to the lifespan of an object person when health-related information of the object person is input, the model being obtained by learning using the health-related information of each of a plurality of people belonging to a first population and first statistical information on the lifespan of the first population; a statistical information acquisition unit configured to acquire the first statistical information of a second population to which the object person belongs; a health-related information acquisition unit configured to acquire the health-related information of the object person; a first type score calculation unit configured to calculate a first type score related to the lifespan of the object person based on the index output as a result of inputting the acquired health-related information into the all-cause mortality model and the acquired first statistical information.

2. The information processing apparatus according to claim 1, wherein the health-related information includes at least one of background information, health examination information, risk factor information, lifestyle information, and family history information.

3. The information processing apparatus according to claim 2, wherein the first statistical information at least includes information on a life table of lifespan.

4. The information processing apparatus according to claim 3, wherein the index related to the lifespan is the position of the object person in the population estimated lifespan distribution, and the first type score calculation unit calculates a score related to the lifespan of the object person as the first type score based on the life table of the second population and the position of the object person in the population estimated lifespan distribution.

5. The information processing apparatus according to claim 4, wherein the first type score calculation unit predicts the remaining lifespan of the object person based on the life table of the second population and the position of the object person in the population estimated lifespan distribution, and calculates a value obtained by dividing the predicted remaining lifespan of the object person by the average lifespan of people of the same age as the object person in the second population as the first type score.

6. The information processing apparatus according to claim 1, wherein, Further comprising: an information providing control unit configured to execute control for providing information including the first type score to a terminal of the object person.

7. The information processing apparatus according to claim 6, wherein, Further comprising: the model acquisition unit is further configured to acquire, as a health condition model, a model that outputs an index related to a predetermined health condition of the object person when the health-related information of the object person is input, the model being obtained by learning using the health-related information of each of the plurality of people belonging to the first population and second statistical information on the predetermined health condition of the first population; the statistical information acquisition unit is further configured to acquire the second statistical information of the second population to which the object person belongs; a second type score calculation unit configured to calculate a second type score related to the predetermined health condition of the object person based on the index output as a result of inputting the health-related information of the object person acquired by the health-related information acquisition unit into the health condition model and the acquired second statistical information. The information providing control unit also executes control to provide the information including the second type of score to the terminal of the target person.

8. The information processing apparatus according to claim 7, wherein the predetermined health condition is one or more health conditions among a disease, a concept explanation, and a composite concept, the concept explanation and the composite concept being scores expressing a health condition and a health awareness, the concept explanation being a score explained according to a predetermined concept, and the composite concept being a score combining a plurality of factors, the second type of score includes scores related to each of the one or more health conditions.

9. The information processing apparatus according to claim 8, wherein, Further comprising: an improvement activity information generation unit that generates information related to an improvement activity based on at least a part of the output results of the all-cause mortality model and the health condition model, the improvement activity indicating an activity that the target person should perform to improve the first type of score, The information providing control unit also executes control to provide the information including the information related to the improvement activity to the terminal of the target person.

10. An information processing method, which is executed by an information processing device, wherein, Comprising: a model acquisition step of acquiring, as an all-cause mortality model, a model that outputs an index related to the life span of a target person when health-related information of the target person is input, the model being obtained by learning using the health-related information of each of a plurality of persons belonging to a first population and first statistical information on the life span of the first population; a statistical information acquisition step of acquiring the first statistical information of the second population to which the target person belongs; a health-related information acquisition step of acquiring the health-related information of the target person; a first type of score calculation step of calculating a first type of score related to the life span of the target person based on the index output as a result of inputting the acquired health-related information into the all-cause mortality model and the acquired first statistical information.

11. A computer-readable recording medium storing a program for causing a computer to execute control processing, wherein, The control process includes: a model acquisition step of acquiring, as an all-cause mortality model, a model that outputs an index related to the life span of a target person when health-related information of the target person is input, the model being obtained by learning using the health-related information of each of a plurality of persons belonging to a first population and first statistical information on the life span of the first population; a statistical information acquisition step of acquiring the first statistical information of the second population to which the target person belongs; a health-related information acquisition step of acquiring the health-related information of the target person; a first type of score calculation step of calculating a first type of score related to the life span of the target person based on the index output as a result of inputting the acquired health-related information into the all-cause mortality model and the acquired first statistical information.

Images