Information processing system, information processing method and program

Abstract

To determine a mild cognitive impairment or its risk highly accurately.SOLUTION: An information processing system has a control unit. The control unit is configured to let level information on D-amino acid in a bio-sample collected respectively from a healthy subject, and a subject who is evaluated that the subject could suffer from at least a mild cognitive impairment on the basis of a prescribed examination, and cohabitant composition information indicative of a cohabitant composition of each subject be an explanatory variable; input level information on the D-amino acid in the bio sample collected from an arbitrary user, and cohabitant composition information indicative of a cohabitant composition of the user into a learned model generated by learning to predict presence or absence of the mild cognitive impairment of each subject, or its risk thereof; and acquire determination information on the presence or absence of the mild cognitive impairment or its risk of the user predicted by the learned model, from the learned model.SELECTED DRAWING: Figure 6

Description

[Technical Field]

[0001] The present invention relates to an information processing system, an information processing method, and a program capable of processing information for determining a user's mild cognitive impairment or the risk thereof. [Background technology]

[0002] It has long been known that a person's cognitive function is related to the level (ratio) of D-amino acids that the person possesses. Patent Document 1 below discloses a testing device for mild cognitive impairment or dementia or the risk thereof, which includes a means for measuring the amount of amino acid stereoisomers in a biological sample, a means for comparing the D-amino acid level with a reference value, and a means for outputting information on the subject's pathology or risk thereof based on the comparison. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2020-190542 Summary of the Invention [Problem to be solved by the invention]

[0004] An object of the present invention is to provide an information processing system, an information processing method, and a program that are capable of determining mild cognitive impairment or the risk thereof with higher accuracy. [Means for solving the problem]

[0005] An information processing system according to one embodiment of the present invention includes a control unit. The control unit inputs D-amino acid level information in a biological sample collected from a user and the cohabitation information indicating the cohabitation of the user into a trained model generated by training the model to predict whether or not each subject has mild cognitive impairment or is at risk for mild cognitive impairment, using D-amino acid level information in biological samples collected from healthy subjects and subjects evaluated as suspected of having at least mild cognitive impairment through a predetermined test and cohabitation information indicating the cohabitation of the user as explanatory variables. The control unit then obtains, from the trained model, information on the presence or absence of mild cognitive impairment or the risk of mild cognitive impairment predicted by the trained model.

[0006] Another aspect of the present invention provides an information processing system including a control unit. The control unit acquires D-amino acid level information from biological samples collected from healthy subjects and subjects who have been evaluated by a predetermined test to be at least suspected of having mild cognitive impairment, as well as housemate composition information indicating the housemate composition of each subject. The control unit then uses the level information and the housemate composition information as explanatory variables and performs training to predict the mild cognitive impairment or risk of mild cognitive impairment for each subject, thereby generating a trained model that determines the presence or absence of mild cognitive impairment or the risk of mild cognitive impairment for a given user based on the level information and the housemate composition information for the user.

[0007] An information processing method according to another aspect of the present invention includes: D-amino acid level information in biological samples collected from a user and the cohabitation information indicating the cohabitation composition of the user are input into a trained model generated by performing training to predict the presence or absence of mild cognitive impairment or a risk thereof for each of the subjects, using as explanatory variables information on D-amino acid levels in biological samples collected from each of healthy subjects and subjects who have been evaluated as being at least suspected of having mild cognitive impairment by a predetermined test, and information on the cohabitation composition indicating the cohabitation composition of the user; This includes obtaining, from the trained model, judgment information regarding the presence or absence of mild cognitive impairment or the risk of mild cognitive impairment in the user, as predicted by the trained model.

[0008] An information processing method according to yet another aspect of the present invention includes: Acquiring information on D-amino acid levels in biological samples collected from healthy subjects and subjects who have been assessed by a predetermined test as being suspected of having at least mild cognitive impairment, and information on the cohabitant composition of each of the subjects; The method includes generating a trained model that determines whether or not a user has mild cognitive impairment or the risk of mild cognitive impairment based on the level information and the cohabitant composition information of any user, by using the level information and the cohabitant composition information as explanatory variables and learning to predict the mild cognitive impairment or the risk of mild cognitive impairment for each subject.

[0009] According to yet another aspect of the present invention, there is provided a program for executing the program on an information processing device, inputting D-amino acid level information in biological samples collected from any user and the cohabitation information indicating the cohabitation composition of the user into a trained model generated by learning to predict the presence or absence of mild cognitive impairment or its risk in each subject, using D-amino acid level information in biological samples collected from each healthy subject and a subject who has been assessed as suspected of having at least mild cognitive impairment by a predetermined test, and the cohabitation information indicating the cohabitation composition of the subject, as explanatory variables; The system also executes a step of obtaining, from the trained model, judgment information on whether the user has mild cognitive impairment or is at risk of having mild cognitive impairment, as predicted by the trained model.

[0010] According to yet another aspect of the present invention, there is provided a program for executing the program on an information processing device, acquiring information on D-amino acid levels in biological samples collected from healthy subjects and subjects who have been assessed by a predetermined test as being suspected of having at least mild cognitive impairment, and information on the cohabitant composition of each of the subjects; The method executes a step of generating a trained model that determines whether or not a user has mild cognitive impairment or the risk of mild cognitive impairment based on the level information and the cohabitant composition information of any user, by using the level information and the cohabitant composition information as explanatory variables and performing learning to predict the mild cognitive impairment or the risk of mild cognitive impairment for each subject. [Effects of the Invention]

[0011] According to an information processing system according to an embodiment of the present invention, mild cognitive impairment or the risk thereof can be determined with high accuracy. However, this effect does not limit the present invention. [Brief explanation of the drawings]

[0012] [Figure 1] 1 is a diagram showing the configuration of a mild cognitive impairment assessment system according to one embodiment of the present invention. [Figure 2] 1 is a diagram showing the hardware configuration of a mild cognitive impairment assessment server according to one embodiment of the present invention. FIG. [Figure 3] 1 is a diagram showing the configuration of a database possessed by a mild cognitive impairment assessment server according to one embodiment of the present invention. FIG. [Figure 4] 10 is a flowchart showing the flow of a process for generating a mild cognitive impairment prediction model by a mild cognitive impairment assessment server according to one embodiment of the present invention. [Figure 5] 1 is a diagram illustrating the prediction accuracy of a mild cognitive impairment prediction model constructed by a mild cognitive impairment assessment server according to one embodiment of the present invention. FIG. [Figure 6] 10 is a flowchart showing the flow of a mild cognitive impairment assessment process using the mild cognitive impairment prediction model by a mild cognitive impairment assessment server according to one embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0013] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0014] [System Configuration] As shown in Figure 1, the mild cognitive impairment assessment system of this embodiment includes a mild cognitive impairment assessment server 100 on the Internet 50, a plurality of user terminals 200, and a testing institution terminal 300 of a testing institution (including a medical institution).

[0015] The mild cognitive impairment assessment server 100 provides a service for assessing mild cognitive impairment (MCI) or its risk to users of user terminals 200. The mild cognitive impairment assessment server 100 (hereinafter also referred to as the "MCI assessment server 100") is connected to a plurality of user terminals 200 and a testing institution terminal 300 of a testing institution via the Internet 50.

[0016] The MCI assessment server 100 assesses the user's MCI or risk of MCI based on D-amino acid level (ratio) information extracted from a biological sample collected from the user of the user terminal 200 and the response information to a questionnaire obtained from the user, and transmits the assessment information to the user terminal 200.

[0017] The user terminals 200 (200A, 200B, 200C, etc.) are terminals used by users, such as smartphones, mobile phones, tablet PCs (Personal Computers), notebook PCs, and desktop PCs. The user terminals 200 transmit an MCI assessment request together with questionnaire response information to the MCI assessment server 100, receive MCI assessment information from the MCI assessment server 100, and display the information on a screen using a browser or the like. An application compatible with the MCI assessment service (hereinafter also referred to as an MCI app) may be installed in the user terminals 200, and the user terminals 200 may transmit an MCI assessment request to the MCI assessment server 100 and display the MCI assessment information using the MCI app.

[0018] Testing institution terminal 300 analyzes the user's biological sample, extracts D-amino acid level (ratio) information (hereinafter simply referred to as "D-amino acid information"), and transmits the D-amino acid information to MCI assessment server 100. If user terminal 200 stores D-amino acid information previously extracted from a biological sample at a testing institution or the like, MCI assessment server 100 may receive the D-amino acid information from user terminal 200. Alternatively, the testing institution may administer a questionnaire to the user, and the resulting questionnaire information may be transmitted from testing institution terminal 300 to MCI assessment server 100.

[0019] The biological sample may be collected by the user using a collection kit provided to the user in advance and mailed to a testing institution, or may be collected by the user visiting a testing institution. The testing institution terminal 300 may be a terminal of an organization operated by an operator different from that of the MCI assessment server 100, or may be a terminal of the same organization as the operator of the MCI assessment server 100. The collection kit may also be sold to companies other than the operator or to testing institutions.

[0020] The biological sample is, for example, a blood sample, more specifically, a fingertip blood sample, but instead of this, a venous blood sample from the forearm or the like may be used, and other biological samples such as urine, saliva, sebum, and stratum corneum may be used in addition to a blood sample.

[0021] The level (ratio) information of D-amino acids is, for example, information on the levels of D-alanine, D-proline, and D-serine in a biological sample, specifically, the chiral balance of each of D-alanine, D-proline, and D-serine. The level of D-alanine is expressed as [D-alanine amount / (D-alanine amount+L-alanine amount)]×100, the level of D-proline is expressed as [D-proline amount / (D-proline amount+L-proline amount)]×100, and the level of D-serine is expressed as [D-alanine amount / (D-alanine amount+L-alanine amount)]×100. However, without being limited thereto, for example, the chiral balance of any one or two of D-alanine, D-proline, and D-serine may be used as the level information.

[0022] It has long been known that D-amino acid levels are involved in human cognitive function, but the inventors attempted to further improve the accuracy of predicting MCI or its risk by selecting specific information from various lifestyle information, physical information, etc. and adding it to the explanatory variables in addition to the D-amino acid level information.

[0023] To assess MCI using the D-amino acid information and questionnaire response information (hereinafter simply referred to as "questionnaire information"), the MCI assessment server 100 uses an MCI assessment model 10. The MCI assessment model 10 is a trained model that uses the questionnaire information and D-amino acid level information in the biological samples collected from healthy subjects and subjects who have been assessed as at least suspected of having MCI through a predetermined test as explanatory variables, and learns to predict MCI or the risk of MCI for each subject, thereby determining whether or not a user has MCI or is at risk for MCI based on the user's D-amino acid level information and questionnaire information. Details of the process of constructing the MCI assessment model 10 and the MCI assessment process using it will be described later.

[0024] [Hardware configuration of the mild cognitive impairment assessment server] FIG. 2 is a diagram showing the hardware configuration of the mild cognitive impairment assessment server 100. As shown in FIG.

[0025] As shown in the figure, the MCI assessment server 100 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, an input / output interface 15, and a bus 14 connecting these components to one another.

[0026] The CPU 11 accesses the RAM 13 etc. as needed, and performs various arithmetic processing while comprehensively controlling each block of the mild cognitive impairment assessment server 100. Multiple CPUs 11 may be provided depending on the processing. The ROM 12 is a non-volatile memory in which firmware such as the OS, programs, and various parameters to be executed by the CPU 11 are permanently stored. The RAM 13 is used as a working area for the CPU 11, and temporarily stores the OS, various applications currently being executed, and various data currently being processed.

[0027] The input / output interface 15 is connected to a display unit 16, an operation reception unit 17, a storage unit 18, a communication unit 19, and the like.

[0028] The display unit 16 is a display device that uses, for example, an LCD (Liquid Crystal Display), an OLED (Organic ElectroLuminescence Display), a CRT (Cathode Ray Tube), or the like.

[0029] The operation reception unit 17 is, for example, a pointing device such as a mouse, a keyboard, a touch panel, or other input device. When the operation reception unit 17 is a touch panel, the touch panel can be integrated with the display unit 16.

[0030] The storage unit 18 is a non-volatile memory such as a hard disk drive (HDD), a flash memory (solid state drive (SSD)), or other solid-state memory. The storage unit 18 stores the OS, various applications, and various data.

[0031] As will be described later, particularly in this embodiment, the memory unit 18 has programs such as applications necessary for the generation process of the MCI determination model 10 described later and the MCI determination process using the MCI determination model 10, as well as a D-amino acid information database, a questionnaire information database, and an MCI determination information database.

[0032] The communication unit 19 is, for example, a NIC (Network Interface Card) for Ethernet or various modules for wireless communication such as wireless LAN, and is responsible for communication processing between the user terminal 200 and the inspection agency terminal 300.

[0033] Although not shown, the basic hardware configuration of the user terminal 200 is also substantially the same as the hardware configuration of the MCI assessment server 100 described above.

[0034] [Mild cognitive impairment assessment server database configuration]

[0035] 3, the MCI assessment server 100 has, in the storage unit 18, a D-amino acid information database 31, a questionnaire information database 32, and an MCI assessment information database 33. Note that these databases may be stored in a storage device or server externally connected to the MCI assessment server 100, rather than in the storage unit 18.

[0036] The D-amino acid information database 31 stores the user's D-amino acid information received from the testing institution terminal 300 in association with general information such as the user's name, user ID for identifying the user, date of birth, occupation, address, telephone number, and email address.

[0037] The questionnaire information database 32 stores information on responses to questionnaires received from the user terminal 200. In this embodiment, the questionnaire information includes at least information on the user's cohabitants, the number of cohabitants, and their ages. Cohabitants may or may not be related by blood. Specifically, the cohabitant information includes, but is not limited to, information on spouses, common-law spouses, parents, children, and friends.

[0038] The above questionnaire information (information on cohabitant composition, number of cohabitants, and age information) may be generated by dialogue with the user or by user input at the testing institution that conducted the testing of the above biological sample, and may be transmitted from the testing institution terminal 300 to the MCI assessment server 100.

[0039] The MCI assessment information database 33 stores MCI assessment information, which is the result of assessment by the MCI assessment model 10 based on the D-amino acid information and questionnaire information, in association with a user ID. The MCI assessment information includes, for example, MMSE (Mini-Mental State Examination) score information and corresponding assessment information indicating either MCI (suspected) or healthy status; however, for example, only the MMSE score information or only the assessment information of MCI (suspected) or healthy status based on the MMSE score may be the final MCI assessment information.

[0040] These databases are mutually referenced and used as necessary in the process of generating the MCI determination model 10 by the MCI determination server 100 and the MCI determination process using the same, which will be described later.

[0041] [Operation of the mild cognitive impairment assessment server] Next, we will explain the operation of the MCI assessment server 100 configured as above. This operation is performed by the cooperation of hardware such as the CPU 11 and communication unit 19 of the MCI assessment server 100 and software stored in the storage unit 18. For convenience, in the following explanation, the CPU 11 is the subject of the operations.

[0042] (Mild cognitive impairment prediction model generation process) First, the process of generating the MCI determination model 10 by the MCI determination server 100 will be described.

[0043] 4 is a flowchart showing details of the process of generating the MCI determination model 10. In generating the MCI determination model 10, the inventors used 100 healthy subjects with an MMSE score of 27 or more and 100 MCI suspects aged 50 to 89 with an MMSE score of 26 or less as subjects.

[0044] As shown in the figure, CPU 11 first acquires the subject's D-amino acid information and questionnaire information from D-amino acid information database 31 and questionnaire information database 32 (step 41). As described above, the D-amino acid information is previously received by MCI assessment server 100 from testing institution terminal 300 and stored in D-amino acid information database 31. The questionnaire information is received from user terminal 200 via a questionnaire (questionnaire) screen displayed on user terminal 200 by the MCI app or the like, which has questions about cohabitation, number of cohabitants, and ages, along with corresponding answer fields, and is stored in questionnaire information database 32.

[0045] Next, the CPU 11 performs pre-processing on each piece of received information, that is, standardization processing by setting the average value to 0 and the standard deviation to 1 (step 42).

[0046] Next, the CPU 11 divides the standardized data into training (teacher) data and verification data (step 43).

[0047] Next, the CPU 11 performs nonlinear classification using the feature quantities of the training data as training data (explanatory variables) to construct a model for determining the presence or absence of MCI (for predicting the MMSE score) (step 44).

[0048] The CPU 11 then inputs the validation data into the constructed determination model to determine the presence or absence of MCI (MMSE score), and performs leave-one-out cross validation (LOOCV) by repeating this determination process between data from multiple subjects, changing the training data and validation data (step 45), and verifies the prediction accuracy (step 46).Finally, the CPU 11 uses all the data to construct a final determination model for determining the presence or absence of MCI (MMSE score) for any user.

[0049] 5 is a diagram showing the determination accuracy of the MCI determination model, along with a comparative example. The inventors randomly divided the training data and validation data into three parts, and verified the LOOCV accuracy for each part based on the AUC value.

[0050] As shown in the figure, when, in addition to D-amino acid information, questionnaire information including information on cohabitant composition, number of cohabitants, and age information was learned as explanatory variables, the MCI assessment model constructed from this information achieved an average AUC value of 0.789 over three runs, which was significantly higher in assessment accuracy than when only D-amino acid information was used as an explanatory variable (0.666).

[0051] It was also found that when information on the composition of people living with others was used as an explanatory variable in the questionnaire information, the accuracy of the judgment was higher than when it was not used.It was also found that higher accuracy of the judgment could be achieved by using gender information, years of education information, and BMI information as explanatory variables in addition to information on the composition of people living with others, number of people living with others, and age information.

[0052] (Mild cognitive impairment assessment process) Next, a description will be given of a process for providing mild cognitive impairment assessment information to the user terminal 200 using the mild cognitive impairment prediction model constructed above. Fig. 6 is a flowchart showing the flow of the process for providing mild cognitive impairment assessment information.

[0053] As shown in the figure, first, the CPU 11 determines whether or not an MCI determination request has been received from the user terminal 200 via, for example, an MCI application (step 61).

[0054] If it is determined that the MCI assessment request has been received (Yes in step 61), CPU 11 extracts D-amino acid information corresponding to the user ID of user terminal 200 that has made the MCI assessment request from D-amino acid information database 31, and also obtains questionnaire information included in the MCI assessment request and inputs it to MCI assessment model 10 (step 62). The questionnaire information is also stored in questionnaire information database 32.

[0055] Next, the CPU 11 acquires the MCI judgment information (MMSE score information) for the input from the MCI judgment model 10 (step 63). The MCI judgment information is stored in the MCI judgment information database 33.

[0056] Then, the CPU 11 generates final MCI assessment information including the assessment result (healthy / suspected MCI) corresponding to the MMSE score, and transmits it to the user terminal 200 (step 64).

[0057] If the MCI assessment information indicates suspicion of MCI, the CPU 11 may generate advice information recommending a visit to a hospital (neurological medicine / surgery or psychiatry) along with the MCI assessment information, or hyperlink information to information on nearby hospitals based on the user's address information or the location information of the user terminal 200, and transmit this information to the user terminal 200.

[0058] As described above, according to this embodiment, mild cognitive impairment or the risk thereof can be determined with high accuracy, and the user can be made aware of this.

[0059] [Variations] Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments, and it goes without saying that various modifications can be made within the scope of the gist of the present invention.

[0060] In the above-described embodiment, the user's cohabitation number information, cohabitation number information, and age information were used as the explanatory variables of the MCI determination model 10. However, the questionnaire information only needs to include at least cohabitation number information, and information on the number of cohabitation people and their ages is not essential. Two items, cohabitation number information and cohabitation number information, or two items, cohabitation number information and age information, may be used for the determination. In addition to these, at least one of the user's gender information, years of education information, and BMI information may be used as the explanatory variables.

[0061] In the above-described embodiment, the MCI assessment server 100 receives an MCI assessment request from the user terminal 200 and transmits MCI assessment information to the user terminal 200. However, the source of the MCI assessment request and the destination of the MCI assessment information are not limited to the user terminal 200 and may be a third-party terminal. For example, the MCI assessment server 100 may receive a user's MCI assessment request from a testing institution terminal 300 of a testing institution that tested the user's biological sample and extracted D-amino acid information, and transmit MCI assessment information to the testing institution terminal 300 or the user terminal 200 in response. Furthermore, if an MCI assessment service is used as part of a company's employee benefits program or a local government subsidy program, the MCI assessment server 100 may receive a user's MCI assessment request from an organization terminal of the organization, such as the company or local government, to which the user belongs, and transmit MCI assessment information to the organization terminal or the user terminal 200 in response. In this case, the company or local government may purchase a collection kit from the operator of the MCI assessment server 100 or another company and distribute it to the user. Furthermore, a company or local government may conduct a questionnaire with users, generate response information to the questionnaire, and transmit the information to the MCI assessment server 100.

[0062] In the above-described embodiment, the MCI assessment server 100 transmits the generated MCI assessment information to the user terminal 200. However, the transmission process of the MCI assessment information is not essential, and the generated MCI assessment information may simply be stored in the MCI assessment information database 33. Thereafter, the MCI assessment information may be output onto a paper medium and mailed to the user, the testing institution, or the above-described organization.

[0063] Although the above-described embodiment shows only one MCI determination server 100, the processes executed by the MCI determination server 100 may be distributed among a plurality of servers. For example, the process of generating the MCI determination model 10 and the MCI determination process using the MCI determination model 10 may be executed by separate servers. That is, the MCI determination server 100 may not construct the MCI determination model 10 but may only perform the MCI determination process using the MCI determination model, or conversely, may only construct the MCI determination model but not perform the MCI determination process using the MCI determination model.

[0064] Among the inventions described in the claims of this application, the invention described as an "information processing method" is one in which each step is automatically performed by at least one device such as a computer through software-based information processing, and is not performed by a human using a device such as a computer. In other words, the "information processing method" is an information processing method using computer software, and is not a method in which a human operates a computing tool called a computer. [Explanation of symbols]

[0065] 10...Mild cognitive impairment (MCI) assessment model 11...CPU 18...Storage section 19…Communications Department 31...D-amino acid information database 32...Questionnaire information database 33…MCI assessment information database 100...Mild cognitive impairment (MCI) assessment server 200...User terminal 300...Inspection agency terminal

Claims

1. D-amino acid level information in biological samples collected from a user and the cohabitation information indicating the cohabitation composition of the user are input into a trained model generated by performing training to predict the presence or absence of mild cognitive impairment or a risk thereof for each of the subjects, using as explanatory variables information on D-amino acid levels in biological samples collected from a healthy subject and a subject who has been evaluated by a predetermined test as being suspected of having at least mild cognitive impairment, and information on the cohabitation composition indicating the cohabitation composition of the subject; From the trained model, determination information on the presence or absence of mild cognitive impairment or the risk thereof of the user predicted by the trained model is obtained. Control Unit An information processing system comprising:

2. The trained model is generated by further using, as the explanatory variable, information on the number of people living together with each subject, the number of people living together with each subject; The control unit further inputs information about the number of people living together with the user into the trained model. The information processing system according to claim 1 .

3. The trained model is generated by further using age information indicating the age of each subject as an explanatory variable, The control unit further inputs age information of the user into the trained model. The information processing system according to claim 1 .

4. The control unit receives the level information of the user from an institution terminal of the testing institution that collected the biological sample, and receives the cohabitant composition information from the institution terminal or the user terminal of the user. The information processing system according to claim 1 .

5. The control unit transmits the acquired determination information to the user terminal or the organization terminal. The information processing system according to claim 4 .

6. The control unit receives a request to transmit the determination information of the user from a user terminal of the user, an institution terminal of a testing institution that collected the biological sample of the user, or an organization terminal of an organization to which the user belongs, and transmits the acquired determination information to the user terminal, institution terminal, or organization terminal. The information processing system according to claim 1 .

7. Acquiring information on D-amino acid levels in biological samples collected from healthy subjects and subjects who have been assessed by a predetermined test as being suspected of having at least mild cognitive impairment, and information on the cohabitant composition of each of the subjects; The level information and the cohabitant composition information are used as explanatory variables, and learning is performed to predict the mild cognitive impairment or the risk thereof for each of the subjects, thereby generating a trained model that determines whether or not a user has mild cognitive impairment or the risk thereof based on the level information and the cohabitant composition information of the user. Control Unit An information processing system comprising:

8. D-amino acid level information in biological samples collected from a user and the cohabitation information indicating the cohabitation composition of the user are input into a trained model generated by performing training to predict the presence or absence of mild cognitive impairment or a risk thereof for each of the subjects, using as explanatory variables information on D-amino acid levels in biological samples collected from a healthy subject and a subject who has been evaluated by a predetermined test as being suspected of having at least mild cognitive impairment, and information on the cohabitation composition indicating the cohabitation composition of the subject; From the trained model, determination information on the presence or absence of mild cognitive impairment or the risk thereof of the user predicted by the trained model is obtained. Information processing methods.

9. Acquiring information on D-amino acid levels in biological samples collected from healthy subjects and subjects who have been assessed by a predetermined test as being suspected of having at least mild cognitive impairment, and information on the cohabitant composition of each of the subjects; The level information and the cohabitant composition information are used as explanatory variables, and learning is performed to predict the mild cognitive impairment or the risk thereof for each of the subjects, thereby generating a trained model that determines whether or not a user has mild cognitive impairment or the risk thereof based on the level information and the cohabitant composition information of the user. Information processing methods.

10. In the information processing device, inputting D-amino acid level information in biological samples collected from any user and the cohabitation information indicating the cohabitation composition of the user into a trained model generated by learning to predict the presence or absence of mild cognitive impairment or its risk in each of the subjects, using D-amino acid level information in biological samples collected from healthy subjects and subjects who have been evaluated as suspected of having at least mild cognitive impairment by a predetermined test, and the cohabitation information indicating the cohabitation composition of the user, as explanatory variables; A step of obtaining, from the trained model, determination information on the presence or absence of mild cognitive impairment or a risk thereof of the user predicted by the trained model; A program that executes the following.

11. In the information processing device, acquiring information on D-amino acid levels in biological samples collected from healthy subjects and subjects who have been assessed by a predetermined test as being suspected of having at least mild cognitive impairment, and information on the cohabitant composition of each of the subjects; A step of generating a trained model that determines whether or not a user has mild cognitive impairment or a risk thereof from the level information and the cohabitant composition information of an arbitrary user by performing learning to predict the mild cognitive impairment or a risk thereof of each subject using the level information and the cohabitant composition information as explanatory variables; A program that executes the following.

Images