Layering method and layering device

Abstract

Provided is a layering method including: an obtaining step of obtaining, for each of a plurality of classification subjects having mild cognitive impairment, a score of a result of a clinical psychological test performed on the classification subject, and a hippocampus volume of the classification subject; a classification step of classifying each of the plurality of classification objects according to the score of the classification object and the hippocampus volume of the classification object, classifying the object to be classified into one of a plurality of groups including at least a first group in which the risk of developing into Alzheimer's disease in the future is relatively high and a second group in which the risk of developing into Alzheimer's disease in the future is relatively low; and an output step of outputting a classification result obtained through the classification step.

Description

[0001] This application is a divisional application of a Chinese patent application with the application number "202280061563.4", the application date of December 16, 2022, and the title of "Stratification Method and Stratification Device". Technical Field

[0002] The present invention relates to a stratification method for disease risk, which accurately stratifies the risk levels of each of a plurality of classification objects with mild cognitive impairment (MCI) according to their likelihood of developing Alzheimer's disease. Background Art

[0003] Among the mild cognitive impairment population, there are those who will develop Alzheimer's disease in the future and those who will not.

[0004] Among the mild cognitive impairment population, if those with a relatively high risk of developing Alzheimer's disease in the future can be used as research subjects for clinical trials, etc., then such clinical trials can be effectively carried out and effective prevention methods can be established.

[0005] In the prior art, a plurality of classification objects with mild cognitive impairment are respectively classified into one of a plurality of groups corresponding to the risk of developing Alzheimer's disease in the future. For example, there is the technology described in Non-Patent Document 1 which is known.

[0006] (Prior Art Documents)

[0007] (Non-Patent Documents)

[0008] Non-Patent Document 1: Bin Zhou et.al, Protective Factors Modulate the Risk of Beta Amyloid in Alzheimer’s Disease, Behavioral Neurology, Volume 2020, Article ID 7029642 Summary of the Invention

[0009] Problems to be Solved by the Invention

[0010] In the prior art described in Non-Patent Document 1, the above classification is carried out using the results of brain β-amyloid PET examinations.

[0011] The price of brain β-amyloid PET examination is relatively expensive. Therefore, the above classification carried out by the prior art described in Non-Patent Document 1 is difficult to be widely popularized. In addition, in brain β-amyloid PET examination, only about 41% of the people with abnormal examination developed Alzheimer's disease in the follow-up survey of 26 months on average.

[0012] Then, in view of the above problems, the object of the present invention is to provide a stratification method, etc. The present invention adopts a method different from the prior art using the results of brain β-amyloid PET examination, and classifies each individual with mild cognitive impairment into corresponding groups according to the high or low risk of developing Alzheimer's disease in the future.

[0013] Means for solving the problem

[0014] The stratification method according to one aspect of the present invention includes the following steps: a classification step of classifying each of a plurality of classification objects with mild cognitive impairment into one of a plurality of groups including at least a first group with a relatively high risk of developing Alzheimer's disease in the future and a second group with a relatively low risk of developing Alzheimer's disease in the future according to the score of the result of a clinical psychological test for the classification object and the hippocampal volume of the classification object; and an output step of outputting the classification result of the classification step. The respective hippocampal volumes of the plurality of classification objects are the respective left hippocampal volumes of the plurality of classification objects, the clinical psychological test is the Alzheimer's Disease Assessment Scale - Cognitive Subscale. In the classification step, a volume score value is calculated for each of the plurality of classification objects, the volume score value shows the ratio of the left hippocampal volume of the classification object to the score of the classification object, and the classification object is classified into one of the plurality of groups according to the calculated volume score value.

[0015] A stratification device according to one aspect of the present invention includes: an acquisition unit that acquires, for each of a plurality of classification objects with mild cognitive impairment, a score of a result of a clinical psychological test for the classification object and the hippocampal volume of the classification object; a classification unit that classifies, for each of the plurality of classification objects, the classification object into one of a plurality of groups including at least a first group with a relatively high risk of developing Alzheimer's disease in the future and a second group with a relatively low risk of developing Alzheimer's disease in the future, based on the score of the classification object and the hippocampal volume of the classification object; and an output unit that outputs the classification result obtained by the classification unit. The hippocampal volume of each of the plurality of classification objects is the left hippocampal volume of each of the plurality of classification objects, the clinical psychological test is the Alzheimer's Disease Assessment Scale - Cognitive Subscale, the classification unit calculates a volume score value for each of the plurality of classification objects, the volume score value indicates the ratio of the left hippocampal volume of the classification object to the score of the classification object, and the classification object is classified into one of the plurality of groups based on the calculated volume score value.

[0016] Advantages of the Invention

[0017] By the stratification method and the like according to one aspect of the present invention, each of a plurality of classification objects with mild cognitive impairment can be classified into one of a plurality of groups corresponding to the risk of developing Alzheimer's disease in the future by a method different from the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a schematic diagram showing an example of the content of risk assessment of various factors carried out by the inventors of the present invention.

[0019] Figure 2 It is a block diagram showing an example of the configuration of the stratification method according to Embodiment 1.

[0020] Figure 3 It is a Kaplan-Meier curve (Kaplan-Meier survival curve) showing the results of the follow-up survey carried out by the inventors of the present invention.

[0021] Figure 4 It is a Kaplan-Meier curve (Kaplan-Meier survival curve) showing the results of the follow-up survey of each group carried out by the inventors of the present invention.

[0022] Figure 5 It is a flowchart of the first classification process according to Embodiment 1.

[0023] Figure 6 It is a block diagram showing an example of the configuration of the stratification method according to Embodiment 2.

[0024] Figure 7A It is a Kaplan-Meier curve (Kaplan-Meier survival curve) showing the results of the follow-up surveys of each group conducted by the inventors of the present invention.

[0025] Figure 7B It is a Kaplan-Meier curve (Kaplan-Meier survival curve) showing the results of the follow-up surveys of each group conducted by the inventors of the present invention.

[0026] Figure 8A It is a flowchart of the second classification process according to Embodiment 2.

[0027] Figure 8B It is a flowchart of the second classification process according to Embodiment 2.

[0028] Figure 9 It is a block diagram showing an example of the configuration of the stratification method according to Embodiment 3.

[0029] Figure 10A It is a Kaplan-Meier curve (Kaplan-Meier survival curve) showing the results of the follow-up surveys of each group conducted by the inventors of the present invention.

[0030] Figure 10B It is a Kaplan-Meier curve (Kaplan-Meier survival curve) showing the results of the follow-up surveys of each group conducted by the inventors of the present invention.

[0031] Figure 11 It is a flowchart of the third classification process according to Embodiment 3.

[0032] Figure 12 It is a block diagram showing an example of the configuration of the stratification method according to Embodiment 4.

[0033] Figure 13A It is a flowchart of the fourth classification process according to Embodiment 4.

[0034] Figure 13B It is a flowchart of the fourth classification process according to Embodiment 4.

[0035] Figure 14 It is a block diagram showing an example of the configuration of the stratification method according to Embodiment 5.

[0036] Figure 15A It is a Kaplan-Meier curve (Kaplan-Meier survival curve) showing the results of the follow-up surveys of each group conducted by the inventors of the present invention.

[0037] Figure 15BIt is a Kaplan-Meier curve (Kaplan-Meier survival curve) showing the results of the follow-up surveys of each group conducted by the inventors of the present invention.

[0038] Figure 16 It is a flowchart of the fifth classification process according to Embodiment 5.

[0039] Figure 17 It is a schematic diagram showing an example of the content of the risk assessment targeting various factors further conducted by the inventors of the present invention for positive cases of brain β-amyloid PET examination.

[0040] Figure 18 It is a block diagram showing an example of the configuration of the stratification method according to Embodiment 6.

[0041] Figure 19 It is an example of survival prediction by the product-limit method showing the results of the follow-up surveys of each group conducted by the inventors of the present invention.

[0042] Figure 20 It is a flowchart of the sixth classification process according to Embodiment 6.

[0043] Figure 21 It is a block diagram showing an example of the configuration of the stratification method according to Modification 1.

[0044] Figure 22 It is an example of survival prediction by the product-limit method showing the results of the follow-up surveys of each group conducted by the inventors of the present invention.

[0045] Figure 23 It is a flowchart of the seventh classification process according to Modification 1.

[0046] Figure 24 It is a block diagram showing an example of the configuration of the stratification method according to Modification 2.

[0047] Figure 25 It is an example of survival prediction by the product-limit method showing the results of the follow-up surveys of each group conducted by the inventors of the present invention.

[0048] Figure 26 It is a flowchart of the eighth classification process according to Modification 2.

[0049] Figure 27 It is a block diagram showing an example of the configuration of the stratification method according to Modification 3.

[0050] Figure 28 It is an example of survival prediction by the product-limit method showing the results of the follow-up surveys of each group conducted by the inventors of the present invention.

[0051] Figure 29 It is a flowchart of the 9th classification process related to Modification Example 3.

[0052] Figure 30 It is a block diagram showing an example of the configuration of the layering method related to Embodiment 7.

[0053] Figure 31 It is a correspondence table showing the relationship between the volume fraction value (Cog_Vol) and the number of classification objects developing from mild cognitive impairment to Alzheimer's disease.

[0054] Figure 32 It is a correlation diagram showing the relationship between the volume fraction value (Cog_Vol) and the period from mild cognitive impairment until the development of Alzheimer's disease.

[0055] Figure 33 It is a correspondence table showing the relationship between the volume fraction value (Cog_Vol), multiple thresholds, and the number of classification objects developing from mild cognitive impairment to Alzheimer's disease.

[0056] Figure 34 It is a correlation diagram showing the relationship between the volume fraction value (Cog_Vol), the period from mild cognitive impairment until the development of Alzheimer's disease, or the period elapsed without developing into Alzheimer's disease.

[0057] Figure 35 It is a correspondence table showing the relationship between the volume fraction value (Cog_Vol), multiple thresholds, and the change in the ratio of classification objects developing from mild cognitive impairment to Alzheimer's disease.

[0058] Figure 36 It is a correspondence table showing the relationship between the volume fraction value (Cog_Vol), multiple thresholds, and the number of classification objects developing from mild cognitive impairment to Alzheimer's disease.

[0059] Figure 37 It is a correlation diagram showing the relationship between the volume fraction value (Cog_Vol), the period from mild cognitive impairment until the development of Alzheimer's disease, or the period elapsed without developing into Alzheimer's disease.

[0060] Figure 38 It is a correspondence table showing the relationship between the volume fraction value (Cog_Vol), multiple thresholds, and the change in the ratio of classification objects developing from mild cognitive impairment to Alzheimer's disease.

[0061] Figure 39 It is a correspondence table showing the relationship between the volume fraction value (Cog_Vol_ab), multiple thresholds, and the change in the ratio of classification objects developing from mild cognitive impairment to Alzheimer's disease.

[0062] Figure 40 It is a flowchart of the 10th classification process related to Embodiment 7. Specific Embodiment

[0063] (Process of obtaining one form of the present invention)

[0064] In order to obtain a stratification method that can more accurately classify each of a plurality of classification objects with mild cognitive impairment into one of a plurality of groups corresponding to the risk of developing Alzheimer's disease in the future, the inventors of the present invention evaluated the risk of developing Alzheimer's disease from mild cognitive impairment by taking various factors as objects.

[0065] Figure 1 It is a schematic diagram showing an example of the content of the risk assessment taking various factors as objects carried out by the inventors of the present invention.

[0066] In the process of carrying out the risk assessment on various factors, the inventors of the present invention obtained the following finding: by using an index obtained by combining the score obtained from the results of a clinical psychological test on a classification object and the hippocampal volume of the classification object, the risk of developing Alzheimer's disease from mild cognitive impairment can be evaluated more accurately.

[0067] Then, based on the above finding, the inventors of the present invention further conducted research and discussion. Thus, the following stratification method and the like were formed.

[0068] The stratification method according to one aspect of the present invention includes the following steps: an obtaining step of obtaining, for each of a plurality of classification objects with mild cognitive impairment, the score of the results of a clinical psychological test on the classification object and the hippocampal volume of the classification object; a classification step of classifying, for each of the plurality of classification objects, the classification object into one of a plurality of groups including at least a first group with a relatively high risk of developing Alzheimer's disease in the future and a second group with a relatively low risk of developing Alzheimer's disease in the future according to the score of the classification object and the hippocampal volume of the classification object; and an output step of outputting the classification result obtained by the classification step.

[0069] Through the above stratification method, each of a plurality of classification objects with mild cognitive impairment can be classified into one of a plurality of groups corresponding to the risk of developing Alzheimer's disease in the future by a method different from the conventional technology using the results of brain β-amyloid PET examination.

[0070] And it may also be that the hippocampal volume of each of the plurality of classification objects is the right hippocampal volume of each of the plurality of classification objects.

[0071] Alternatively, the clinical psychological test may be the Alzheimer's Disease Assessment Scale - Cognitive Subscale.

[0072] Alternatively, in the classification step, for each of the multiple classification objects, when the score of the classification object is greater than the first threshold and the right hippocampal volume of the classification object is smaller than the second threshold, the classification object is classified into the first group; when the score of the classification object is smaller than the first threshold and the right hippocampal volume of the classification object is greater than the second threshold, the classification object is classified into the second group.

[0073] Alternatively, the first threshold is 20.0 and the second threshold is 3.651 [cm 3 .

[0074] Alternatively, the clinical psychological test may be the comprehensive score of preclinical Alzheimer's cognitive function detection.

[0075] Alternatively, in the classification step, for each of the multiple classification objects, when the score of the classification object is smaller than the third threshold and the right hippocampal volume of the classification object is smaller than the fourth threshold, the classification object is classified into the first group; when the score of the classification object is greater than the third threshold and the right hippocampal volume of the classification object is greater than the fourth threshold, the classification object is classified into the second group.

[0076] Alternatively, the third threshold is 0.111 and the fourth threshold is 3.651 [cm 3 .

[0077] Alternatively, in the obtaining step, for each of the multiple classification objects, information indicating whether the classification object has the apolipoprotein E ε4 gene is further obtained, and in the classification step, for each of the multiple classification objects, the classification object is also classified into one of the multiple groups according to the information.

[0078] Alternatively, the clinical psychological tests may include: Alzheimer's Disease Assessment Scale - Cognitive Subscale, and the comprehensive score of pre - clinical Alzheimer's cognitive function detection. In the obtaining step, for each of the multiple classified objects, information indicating whether the classified object has the apolipoprotein E ε4 gene is further obtained. In the classifying step, for each of the multiple classified objects, (1) when the first score of the classified object is smaller than the fifth threshold, the right hippocampal volume of the classified object is larger than the sixth threshold, the information of the classified object shows that it has the apolipoprotein E ε4 gene, and the second score of the classified object is smaller than the seventh threshold. The first score is the score obtained as the result of the Alzheimer's Disease Assessment Scale - Cognitive Subscale for the classified object, and the second score is the score obtained as the comprehensive result of pre - clinical Alzheimer's cognitive function detection for the classified object; (2) when the first score of the classified object is larger than the fifth threshold, the right hippocampal volume of the classified object is larger than the sixth threshold, the information of the classified object shows that it has the apolipoprotein E ε4 gene, and the second score of the classified object is smaller than the seventh threshold; (3) when the first score of the classified object is larger than the fifth threshold and the right hippocampal volume of the classified object is smaller than the sixth threshold; (4) when the first score of the classified object is smaller than the fifth threshold, the right hippocampal volume of the classified object is smaller than the sixth threshold, the information of the classified object shows that it has the apolipoprotein E ε4 gene, and the second score of the classified object is smaller than the seventh threshold. The situations of (5) to (7) refer to: (5) when the first score of the classified object is smaller than the fifth threshold, the right hippocampal volume of the classified object is larger than the sixth threshold, the information of the classified object shows that it has the apolipoprotein E ε4 gene, and the second score of the classified object is larger than the seventh threshold; (6) when the first score of the classified object is larger than the fifth threshold, the right hippocampal volume of the classified object is larger than the sixth threshold, the information of the classified object shows that it has the apolipoprotein E ε4 gene, and the second score of the classified object is larger than the seventh threshold; (7) when the first score of the classified object is smaller than the fifth threshold, the right hippocampal volume of the classified object is larger than the sixth threshold, and the information of the classified object shows that it does not have the apolipoprotein E ε4 gene.

[0079] Alternatively, the fifth threshold is 20.0, the sixth threshold is 3.651 [cm 3 , and the seventh threshold is 0.111.

[0080] Alternatively, the multiple classification objects may be positive cases in a brain β-amyloid PET examination, the respective hippocampal volumes of the multiple classification objects may be their respective left hippocampal volumes, and the clinical psychological test may be the Alzheimer's Disease Assessment Scale - Cognitive Subscale.

[0081] Alternatively, in the classification step, for each of the multiple classification objects, if the score of the classification object is greater than the first threshold and the left hippocampal volume of the classification object is less than the second threshold, the classification object is classified into the first group; if the score of the classification object is less than the first threshold and the left hippocampal volume of the classification object is greater than the second threshold, the classification object is classified into the second group.

[0082] Alternatively, the first threshold may be 14.0 and the second threshold may be 3.459 [cm 3 .

[0083] Alternatively, the first threshold may be 12.0 and the second threshold may be 3.737 [cm 3 .

[0084] Alternatively, the first threshold may be 21.0 and the second threshold may be 3.080 [cm 3 .

[0085] Alternatively, the first threshold may be 24.0 and the second threshold may be 2.751 [cm 3 .

[0086] Alternatively, in the classification step, for each of the multiple classification objects, a volume fraction value representing the ratio of the left hippocampal volume of the classification object to the score of the classification object is calculated, and based on the calculated volume fraction value, the classification object is classified into one of the multiple groups.

[0087] Alternatively, the volume fraction value may be the ratio of the left hippocampal volume [mm 3 to the score. In the classification step, for each of the multiple classification objects, if the volume fraction value is less than 101, the classification object is classified into the first group.

[0088] Alternatively, the hippocampal volume of each of the plurality of classification objects may be the left hippocampal volume of each of the plurality of classification objects, the clinical psychological test may be the Alzheimer's Disease Assessment Scale - Cognitive Subscale, and in the classification step, for each of the plurality of classification objects, a volume fraction value representing the ratio of the left hippocampal volume of the classification object to the score of the classification object is calculated, and based on the calculated volume fraction value, the classification object is classified into one of the plurality of groups.

[0089] Alternatively, the volume fraction value may be the ratio of the left hippocampal volume [mm 3 to the score. In the classification step, for each of the plurality of classification objects, when the volume fraction value is less than 101, the classification object is classified into the first group.

[0090] The stratification device according to one aspect of the present invention includes: an acquisition unit that acquires, for each of a plurality of classification objects with mild cognitive impairment, a score of the result of a clinical psychological test for the classification object and the hippocampal volume of the classification object; a classification unit that classifies, for each of the plurality of classification objects, the classification object into one of a plurality of groups including at least a first group with a relatively high risk of developing Alzheimer's disease in the future and a second group with a relatively low risk of developing Alzheimer's disease in the future, based on the score of the classification object and the hippocampal volume of the classification object; and an output unit that outputs the classification result obtained by the classification unit.

[0091] With the above-described stratification device, each of the plurality of classification objects with mild cognitive impairment can be classified into one of a plurality of groups corresponding to the risk of developing Alzheimer's disease in the future by a method different from the conventional technique using the results of brain β-amyloid PET examination.

[0092] Hereinafter, specific examples of the stratification method and the like according to one aspect of the present invention will be described with reference to the drawings.

[0093] The embodiments shown herein are all examples showing the present invention. Therefore, the numerical values, shapes, constituent elements, arrangements and connection manners of the constituent elements, steps (processes), and the order of steps shown in the following embodiments are examples, and the gist thereof is not to limit the present invention. Also, each drawing is a schematic diagram and not a precise illustration. In each drawing, the same reference numerals are given to substantially the same components, and redundant explanations are omitted or simplified.

[0094] In this specification, the stratification method is also referred to as the classification method, and the stratification device is also referred to as the classification device.

[0095] (Embodiment 1)

[0096] The following describes Embodiment 1 of the stratification device for the stratification method involved in Embodiment 1. This stratification device classifies each of a plurality of classification objects with mild cognitive impairment into one of a plurality of groups corresponding to the risk of developing Alzheimer's disease in the future. At least among the plurality of groups, there are a high-risk group (Group 1) with a relatively high risk of developing Alzheimer's disease in the future and a low-risk group (Group 2) with a relatively low risk of developing Alzheimer's disease in the future.

[0097] Moreover, the stratification device that executes the above stratification method is an example of a device that executes classification processing, and in this classification processing, insights obtained by the inventors of the present invention through a cohort study of a plurality of classification objects with mild cognitive impairment are reflected.

[0098] <Configuration>

[0099] Figure 2 It is a block diagram showing an example of the configuration of the stratification device 10 that executes the stratification method involved in Embodiment 1.

[0100] The stratification device 10 is implemented by a computer device, for example. In this case, the stratification device 10 is implemented, for example, by a processor in the computer device executing a program stored in the memory of the computer device.

[0101] As Figure 2 shown, the stratification device 10 includes: an acquisition unit 20, a classification unit 30, and an output unit 40.

[0102] The acquisition unit 20 obtains, for each of a plurality of classification objects with mild cognitive impairment, a score obtained by performing a clinical psychological test on the classification object and the right hippocampal volume of the classification object. The right hippocampal volume can be measured by the following method, for example, by performing an MRI examination on the classification object and performing image processing on the head MRI image obtained as a result of the MRI examination.

[0103] The clinical psychological test to be described in Embodiment 1 for the classification object is the Alzheimer's Disease Assessment Scale - Cognitive Subscale [ADAS-cog (Alzheimer’s Disease Assessment Scale - cognitive subscale)].

[0104] The acquisition unit 20 can, for example, be provided with a communication interface through which scores obtained by performing the Alzheimer's Disease Assessment Scale - Cognitive Subscale on a classification target (hereinafter also referred to as "scores of the Alzheimer's Disease Assessment Scale - Cognitive Subscale") and the right hippocampal volume of the classification target are obtained from an external device. It can also be provided with a memory interface for reading data from a portable and movable memory (for example, a USB memory), and through the memory interface, the scores of the Alzheimer's Disease Assessment Scale - Cognitive Subscale and the right hippocampal volume are obtained from the portable and movable memory. It can also be provided with an input / output interface for accepting operations performed by a user of the hierarchical device 10, and through the input / output interface, a data input operation performed by the user is accepted, and based on this, the scores of the Alzheimer's Disease Assessment Scale - Cognitive Subscale and the right hippocampal volume are obtained.

[0105] For each of a plurality of classification targets, the classification unit 30 classifies the classification target into one of a plurality of groups including at least a first group with a relatively high risk of developing Alzheimer's disease in the future and a second group with a relatively low risk of developing Alzheimer's disease in the future, based on the scores of the Alzheimer's Disease Assessment Scale - Cognitive Subscale and the right hippocampal volume. In Embodiment 1, in addition to the first group and the second group, the plurality of groups further include a third group and a fourth group, and the description is made with the plurality of groups being four groups.

[0106] The classification unit 30 performs the above classification based on the insights obtained by the inventors of the present invention through a cohort study on a plurality of classification targets with mild cognitive impairment as the objects.

[0107] Specifically, for each of a plurality of classification targets, the classification unit 30 performs the following classifications (1) to (4). (1) When the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is greater than the first threshold and the right hippocampal volume is smaller than the second threshold, the classification target is classified into the first group. (2) When the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is smaller than the first threshold and the right hippocampal volume is greater than the second threshold, the classification target is classified into the second group. (3) When the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is greater than the first threshold and the right hippocampal volume is greater than the second threshold, the classification target is classified into the third group. (4) When the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is smaller than the first threshold and the right hippocampal volume is smaller than the second threshold, the classification target is classified into the fourth group. Here, the classification unit 30 sets the first threshold to 20.0 and the second threshold to 3.651 [cm 3 , and performs the above classification.

[0108] The following describes the insights obtained by the inventors of the present invention through a cohort study on a plurality of classification targets with mild cognitive impairment as the objects.

[0109] In a cohort study, the inventors used multiple classified subjects with mild cognitive impairment as the objects and tracked the development from mild cognitive impairment to Alzheimer's disease.

[0110] Figure 3 It is the (Kaplan-Meier survival curve) showing the results of the follow-up investigation conducted by the inventors in the cohort study.

[0111] In Figure 3 In the Kaplan-Meier curve, the horizontal axis is the time [years] elapsed since the start of the follow-up investigation, and the vertical axis is the cumulative conversion rate [%] of the classified subjects who developed from mild cognitive impairment to Alzheimer's disease.

[0112] As Figure 3 shown, in the cohort study, the inventors used 325 classified subjects with mild cognitive impairment as the objects and conducted a follow-up investigation for about 5 years regarding the development from mild cognitive impairment to Alzheimer's disease. During this approximately 5-year period, 104 classified subjects developed from mild cognitive impairment to Alzheimer's disease.

[0113] In the cohort study, for each of the multiple classified subjects, at the start of the follow-up investigation, they were classified into one of the following groups (1) to (4), and the Kaplan-Meier curves showing the results of the follow-up investigation of each classified group were evaluated. The above groups (1) to (4) refer to: (1) Group C (corresponding to Group 1) composed of classified subjects with a score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale greater than the first threshold and a right hippocampal volume smaller than the second threshold, (2) Group B (corresponding to Group 2) composed of classified subjects with a score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale smaller than the first threshold and a right hippocampal volume greater than the second threshold, (3) Group D (corresponding to Group 3) composed of classified subjects with a score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale greater than the first threshold and a right hippocampal volume greater than the second threshold, (4) Group A (corresponding to Group 4) composed of classified subjects with a score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale smaller than the first threshold and a right hippocampal volume smaller than the second threshold.

[0114] Figure 4 It is the Kaplan-Meier curve showing the results of the follow-up investigation for each group conducted by the inventors in the cohort study.

[0115] In Figure 4 In the Kaplan-Meier curve shown, the horizontal axis is the time [years] elapsed since the start of the follow-up investigation, and the vertical axis is the cumulative conversion rate [%] of the classified subjects who developed from mild cognitive impairment to Alzheimer's disease.

[0116] From Figure 4 As can be seen from the Kaplan-Meier curves shown, among the first to fourth groups, the first group has a relatively high risk of developing Alzheimer's disease in the future at the start of the follow-up survey, while the second group has a relatively low risk of developing Alzheimer's disease in the future at the start of the follow-up survey.

[0117] Thus, the inventors of the present invention obtained the following insights through a cohort study. For each of a plurality of classification objects with mild cognitive impairment, (1) when the score on the Alzheimer's Disease Assessment Scale-Cognitive Subscale is greater than the first threshold and the right hippocampal volume is smaller than the second threshold, the classification object is classified into the first group; (2) when the score on the Alzheimer's Disease Assessment Scale-Cognitive Subscale is smaller than the first threshold and the right hippocampal volume is greater than the second threshold, the classification object is classified into the second group; (3) when the score on the Alzheimer's Disease Assessment Scale-Cognitive Subscale is greater than the first threshold and the right hippocampal volume is greater than the second threshold, the classification object is classified into the third group; (4) when the score on the Alzheimer's Disease Assessment Scale-Cognitive Subscale is smaller than the first threshold and the right hippocampal volume is smaller than the second threshold, the classification object is classified into the fourth group. By performing the above classification, a plurality of classification objects with mild cognitive impairment can be classified into one of a plurality of groups (here, four groups) including at least the first group with a relatively high risk of developing Alzheimer's disease in the future and the second group with a relatively low risk of developing Alzheimer's disease in the future.

[0118] Furthermore, the inventors of the present invention obtained the following insights through a cohort study. In the above classification, the first threshold is preferably 20.0 ± 1SD, and the second threshold is preferably 3.651 ± 1SD [cm 3 . Here, 1SD of the first threshold is 6.3, and 1SD of the second threshold is 0.653 [cm 3 .

[0119] Returning again to Figure 2 , the description of the stratifying device 10 will be continued.

[0120] The output unit 40 outputs the classification result performed by the classification unit 30.

[0121] The output unit 40 may, for example, include an output port. Via the output port, a display control signal including the classification result is output to an external display device, and the display control signal is used to cause the external display device to display an image representing the classification result. For example, it may also include a communication interface, and via the communication interface, classification result information including the classification result is transmitted.

[0122] <Operation>

[0123] The operation of the stratifying device 10 having the above configuration will be described below with reference to the drawings.

[0124] Figure 5 It is a flowchart of the first classification process performed by the stratifying device 10. This first classification process is a process of classifying each of a plurality of classification objects into one of four groups including at least a first group with a relatively high risk of developing Alzheimer's disease in the future and a second group with a relatively low risk of developing Alzheimer's disease in the future, based on the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale and the volume of the right hippocampus.

[0125] The first classification process is started, for example, by an operation of a user using the stratifying device 10 to start the first classification process for the stratifying device 10.

[0126] When the first classification process starts, the acquisition unit 20 acquires, for each of a plurality of classification objects with mild cognitive impairment, a score obtained as a result of performing the Alzheimer's Disease Assessment Scale - Cognitive Subscale (ADAS - cog) on the classification object (hereinafter also referred to as the "score of ADAS - cog"), and the volume of the right hippocampus of the classification object (step S5).

[0127] When the scores of ADAS - cog and the volumes of the right hippocampus of a plurality of classification objects are acquired, the classification unit 30 selects one classification object from the plurality of classification objects (step S10). Then, the classification unit 30 investigates whether the score of ADAS - cog of the classification object in the selected state is 20.0 or more (step S15).

[0128] In the process of step S15, when the score of ADAS - cog is 20.0 or more (the "yes" in step S15), the classification unit 30 investigates whether the volume of the right hippocampus of the classification object in the selected state is 3.651 or more (step S20).

[0129] In the process of step S20, when the volume of the right hippocampus is not 3.651 or more (the "no" in step S20), the classification unit 30 classifies the classification object in the selected state into the first group (step S25).

[0130] In the process of step S20, when the volume of the right hippocampus is 3.651 or more (the "yes" in step S20), the classification unit 30 classifies the classification object in the selected state into the third group (step S30).

[0131] In the process of step S15, when the score of ADAS-cog is not 20.0 or higher (the "No" in step S15), the classification unit 30 investigates whether the right hippocampal volume of the classification target in the selected state is 3.651 or higher (step S35).

[0132] In the process of step S35, when the right hippocampal volume is not 3.651 or higher (the "No" in step S35), the classification unit 30 classifies the classification target in the selected state into the 4th group (step S40).

[0133] In the process of step S35, when the right hippocampal volume is 3.651 or higher (the "Yes" in step S35), the classification unit 30 classifies the classification target in the selected state into the 2nd group (step S45).

[0134] When the process of step S25 ends, the process of step S30 ends, the process of step S40 ends, and the process of step S45 ends, the classification unit 30 investigates whether there is an unselected classification target among the multiple classification targets (step S50). Here, the unselected classification target refers to the classification target that was not selected in the process of step S10, and in the loop process that enters step S15 from the process of step S15 until the "Yes" in step S50 and then enters the process of step S15 again, the classification target that has never been selected in the process of step S55 described later.

[0135] In the process of step S50, when there is an unselected classification target (the "Yes" in step S50), the classification unit 30 selects one classification target from the unselected classification targets (step S55).

[0136] When the process of step S55 ends, it enters the process of step S15.

[0137] In the process of step S50, when there is no unselected classification target (the "No" in step S50), the output unit 40 outputs the classification results of each of the multiple classification targets by the classification unit 30 (step S60).

[0138] When the process of step S60 ends, the hierarchical device 10 ends the above-mentioned first classification process.

[0139] <Consideration>

[0140] With the hierarchical device 10 having the above configuration, it is possible to classify multiple classification targets with mild cognitive impairment into one of multiple groups including at least the 1st group with a relatively high risk of developing Alzheimer's disease in the future and the 2nd group with a relatively low risk of developing Alzheimer's disease in the future.

[0141] Therefore, by using the stratification device 10, multiple classification objects with mild cognitive impairment are pre-classified, enabling effective clinical trials and the like with people who develop from mild cognitive impairment to Alzheimer's disease as the research subjects.

[0142] Moreover, with the stratification device 10 having the above configuration, the above classification can be performed by a method different from the prior art. That is, in the above classification by the stratification device 10, the results of the relatively expensive brain β-amyloid PET examination that is necessary in the prior art are not utilized.

[0143] Therefore, by using the stratification device 10, compared with the case of using the prior art, the above classification can be performed inexpensively and with better classification effect.

[0144] (Embodiment 2)

[0145] The stratification device in Embodiment 2 for performing the stratification method according to Embodiment 2 will be described below. In the configuration of Embodiment 2, a part of the functions of the stratification device 10 according to Embodiment 1 is changed. As described in Embodiment 1, the stratification device 10 classifies multiple classification objects with mild cognitive impairment based on the score obtained as the result of the Alzheimer's Disease Assessment Scale - Cognitive Subscale and the right hippocampal volume. In this regard, in the stratification method according to Embodiment 2, in addition to the score obtained as the result of the Alzheimer's Disease Assessment Scale - Cognitive Subscale and the right hippocampal volume, multiple classification objects with mild cognitive impairment are classified based on the presence or absence of the apolipoprotein E ε4 gene. Here, for the stratification device that executes the stratification method according to Embodiment 2, for the same constituent elements as the stratification device 10, since they have been described, the same numbers are given and the detailed description is omitted. The following will focus on the differences from the stratification device 10.

[0146] <Configuration>

[0147] Figure 6 It is a block diagram showing an example of the configuration of the stratification device 10A that executes the stratification method according to Embodiment 2.

[0148] As Figure 6 shown, the difference between the stratification device 10A and the stratification device 10 according to Embodiment 1 is that the acquisition unit 20 is changed to the acquisition unit 20A, and the classification unit 30 is changed to the classification unit 30A.

[0149] For each of a plurality of classified subjects with mild cognitive impairment, the acquisition unit 20A acquires, in addition to the score obtained as the result of the Alzheimer's Disease Assessment Scale - Cognitive Subscale for the classified subject and the right hippocampal volume of the classified subject, information indicating whether the classified subject has the apolipoprotein E ε4 gene (hereinafter also referred to as "ApoE genotyping").

[0150] For each of the plurality of classified subjects, the classification unit 30A classifies the classified subject into one of a plurality of groups including at least a first group with a relatively high risk of developing Alzheimer's disease in the future and a second group with a relatively low risk of developing Alzheimer's disease in the future, based on the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale, the right hippocampal volume, and the ApoE genotyping. In Embodiment 2, the plurality of groups include, in addition to the first group and the second group, a third group, a fourth group, a fifth group, a sixth group, a seventh group, and an eighth group. The following describes the plurality of groups composed of these eight groups.

[0151] The classification unit 30A performs the above classification based on the insights obtained by the inventors of the present invention through a cohort study on a plurality of classified subjects with mild cognitive impairment.

[0152] More specifically, the classification unit 30A classifies each of a plurality of classification targets according to the following (1) to (8). (1) When the ApoE genotyping shows that the classification target has the apolipoprotein E ε4 gene (hereinafter also referred to as "ApoE+"), the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is greater than the first threshold, and the right hippocampal volume is smaller than the second threshold, the classification target is classified into the first group. (2) When the ApoE genotyping shows ApoE+, the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is smaller than the first threshold, and the right hippocampal volume is greater than the second threshold, the classification target is classified into the second group. (3) When the ApoE genotyping shows ApoE+, the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is greater than the first threshold, and the right hippocampal volume is greater than the second threshold, the classification target is classified into the third group. (4) When the ApoE genotyping shows ApoE+, the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is smaller than the first threshold, and the right hippocampal volume is smaller than the second threshold, the classification target is classified into the fourth group. (5) When the ApoE genotyping shows that the classification target does not have the apolipoprotein E ε4 gene (hereinafter also referred to as "ApoE-"), the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is greater than the first threshold, and the right hippocampal volume is smaller than the second threshold, the classification target is classified into the fifth group. (6) When the ApoE genotyping shows ApoE-, the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is smaller than the first threshold, and the right hippocampal volume is greater than the second threshold, the classification target is classified into the sixth group. (7) When the ApoE genotyping shows ApoE-, the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is greater than the first threshold, and the right hippocampal volume is greater than the second threshold, the classification target is classified into the seventh group. (8) When the ApoE genotyping shows ApoE-, the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is smaller than the first threshold, and the right hippocampal volume is smaller than the second threshold, the classification target is classified into the eighth group. Here, the classification unit 30A sets the first threshold to 20.0 and the second threshold to 3.651 [cm 3 , and performs the above classification.

[0153] The insights obtained by the inventors of the present invention through a cohort study of a plurality of classification targets with mild cognitive impairment will be described below.

[0154] In the cohort study, for each of a plurality of classification objects, at the start of the follow-up survey, first, as a first-stage classification, it is classified into (A) group α composed of classification objects without the apolipoprotein E ε4 gene and (B) group β composed of classification objects with the apolipoprotein E ε4 gene. Then, as a second-stage classification, for each of group α and group β, it is classified into one of the following groups (1) to (4), and the Kaplan-Meier curves showing the results of the follow-up surveys of each classified group are evaluated. The classified groups (1) to (4) are: (1) group C composed of classification objects with a score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale greater than the first threshold and a right hippocampal volume smaller than the second threshold (corresponding to the first group in group α and the fifth group in group β); (2) group B composed of classification objects with a score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale smaller than the first threshold and a right hippocampal volume greater than the second threshold (corresponding to the second group in group α and the sixth group in group β); (3) group D composed of classification objects with a score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale greater than the first threshold and a right hippocampal volume greater than the second threshold (corresponding to the third group in group α and the seventh group in group β); (4) group A composed of classification objects with a score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale smaller than the first threshold and a right hippocampal volume smaller than the second threshold (corresponding to the fourth group in group α and the eighth group in group β).

[0155] Figure 7A and Figure 7B are the Kaplan-Meier curves showing the results of the follow-up surveys of each group conducted by the inventors in the cohort study.

[0156] Figure 7A is the Kaplan-Meier curve showing the results of the follow-up surveys of each group conducted with group α as the object, Figure 7B is the Kaplan-Meier curve showing the results of the follow-up surveys of each group conducted with group β as the object.

[0157] In Figure 7A and Figure 7B In the Kaplan-Meier curves shown, the horizontal axis is the time [years] elapsed since the start of the follow-up survey, and the vertical axis is the cumulative conversion rate [%] of classification objects developing from mild cognitive impairment to Alzheimer's disease.

[0158] From Figure 7A and Figure 7BAs can be seen from the Kaplan-Meier curves shown, among the first to eighth groups, the first group is the group with a relatively high risk of developing Alzheimer's disease in the future at the start of the follow-up survey, and the second group is the group with a relatively low risk of developing Alzheimer's disease in the future at the start of the follow-up survey.

[0159] Thus, the inventors of the present invention obtained the following insights through a cohort study. For each of a plurality of classification subjects with mild cognitive impairment, the following classifications (1) to (8) are performed: (1) When the ApoE genotyping shows ApoE+, the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is greater than the first threshold, and the right hippocampal volume is smaller than the second threshold, the classification subject is classified into the first group; (2) When the ApoE genotyping shows ApoE+, the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is smaller than the first threshold, and the right hippocampal volume is greater than the second threshold, the classification subject is classified into the second group; (3) When the ApoE genotyping shows ApoE+, the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is greater than the first threshold, and the right hippocampal volume is greater than the second threshold, the classification subject is classified into the third group; (4) When the ApoE genotyping shows ApoE+, the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is smaller than the first threshold, and the right hippocampal volume is smaller than the second threshold, the classification subject is classified into the fourth group; (5) When the ApoE genotyping shows that the classification subject does not have the apolipoprotein E ε4 gene (hereinafter also referred to as "ApoE-"), the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is greater than the first threshold, and the right hippocampal volume is smaller than the second threshold, the classification subject is classified into the fifth group; (6) When the ApoE genotyping shows ApoE-, the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is smaller than the first threshold, and the right hippocampal volume is greater than the second threshold, the classification subject is classified into the sixth group; (7) When the ApoE genotyping shows ApoE-, the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is greater than the first threshold, and the right hippocampal volume is greater than the second threshold, the classification subject is classified into the seventh group; (8) When the ApoE genotyping shows ApoE-, the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is smaller than the first threshold, and the right hippocampal volume is smaller than the second threshold, the classification subject is classified into the eighth group. By performing the above classifications, a plurality of classification subjects with mild cognitive impairment can be classified into one of a plurality of groups (here, eight groups) including at least the first group with a relatively high risk of developing Alzheimer's disease in the future and the second group with a relatively low risk of developing Alzheimer's disease in the future.

[0160] And, from Figure 7A and Figure 7BAs can be seen from the Kaplan-Meier curves shown, compared with the classification result obtained by performing the above-described second-stage classification on group β composed of classification objects having the apolipoprotein E ε4 gene, the classification result obtained by performing the above-described second-stage classification on group α composed of classification objects not having the apolipoprotein E ε4 gene can more accurately classify multiple classification objects having mild cognitive impairment into one of multiple groups including at least a group with a relatively high risk of developing Alzheimer's disease in the future and a group with a relatively low risk of developing Alzheimer's disease in the future.

[0161] Thus, through the cohort study, the present inventors obtained the following insight: that is, in addition to the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale and the right hippocampal volume, by also considering the ApoE genotype, it is possible to more accurately classify classification objects into one of multiple groups including at least a first group with a relatively high risk of developing Alzheimer's disease in the future and a second group with a relatively low risk of developing Alzheimer's disease in the future.

[0162] <Work>

[0163] The operation of the stratifying device 10A having the above-described configuration will be described below with reference to the drawings.

[0164] The stratifying device 10A performs a second classification process, which is a process obtained by modifying a part of the first classification process performed by the stratifying device 10.

[0165] Figure 8A And Figure 8B is a flowchart of the second classification process performed by the stratifying device 10A.

[0166] As Figure 8A And Figure 8B As shown, compared with the first classification process, the second classification process adds the processes of step S100 to step S180.

[0167] Therefore, the processes of step S100 to step S180 will be mainly described below.

[0168] At the start of the second classification process, the acquisition unit 20A acquires the score of ADAS-cog, the right hippocampal volume of the classification object, and the ApoE genotype for each of multiple classification objects having mild cognitive impairment (step S100).

[0169] At the end of the process of step S100, the process proceeds to step S10.

[0170] When the processing in step S10 ends and the processing in step S55 ends, the classification unit 30A investigates whether the ApoE genotype of the classification target in the selected state indicates ApoE+ (step S110).

[0171] In the processing of step S110, when the ApoE genotype of the classification target in the selected state does not show ApoE+ (the "no" in step S110), the process proceeds to the processing of step S15.

[0172] In the processing of step S110, when the ApoE genotype of the classification target in the selected state shows ApoE+ (the "yes" in step S110), the classification unit 30A investigates whether the score of ADAS-cog of the classification target in the selected state is 20.0 or more (step S120).

[0173] In the processing of step S120, when the score of ADAS-cog is 20.0 or more (the "yes" in step S120), the classification unit 30A investigates whether the right hippocampal volume of the classification target in the selected state is 3.651 or more (step S130).

[0174] In the processing of step S130, when the right hippocampal volume is not 3.651 or more (the "no" in step S130), the classification unit 30A classifies the classification target in the selected state into group 5 (step S140).

[0175] In the processing of step S130, when the right hippocampal volume is 3.651 or more (the "yes" in step S130), the classification unit 30A classifies the classification target in the selected state into group 7 (step S150).

[0176] In the processing of step S120, when the score of ADAS-cog is not 20.0 or more (the "no" in step S120), the classification unit 30A investigates whether the right hippocampal volume of the classification target in the selected state is 3.651 or more (step S160).

[0177] In the processing of step S160, when the right hippocampal volume is not 3.651 or more (the "no" in step S160), the classification unit 30A classifies the classification target in the selected state into group 8 (step S170).

[0178] In the processing of step S160, when the right hippocampal volume is 3.651 or more (the "yes" in step S160), the classification unit 30A classifies the classification target in the selected state into group 6 (step S180).

[0179] When the processing in step S140 ends, the processing in step S150 ends, the processing in step S170 ends, and the processing in step S180 ends, the process proceeds to the processing in step S50.

[0180] When the processing in step S60 ends, the layering device 10A ends this second classification process.

[0181] <Examination>

[0182] The layering device 10A with the above configuration classifies a plurality of classification objects with mild cognitive impairment based on, in addition to the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale and the right hippocampal volume, whether they have the apolipoprotein E ε4 gene. Accordingly, the layering device 10A can classify the classification objects more accurately.

[0183] (Embodiment 3)

[0184] The layering device in Embodiment 3 that executes the layering method according to Embodiment 3 will be described below. The configuration of the layering device according to Embodiment 3 is such that a part of the functions of the layering device 10 according to Embodiment 1 is changed. As described in the explanation of Embodiment 1, the layering device 10 classifies a plurality of classification objects with mild cognitive impairment based on the score obtained from the Alzheimer's Disease Assessment Scale - Cognitive Subscale, which is a clinical psychological test performed on the classification objects, and the right hippocampal volume. Here, the layering method according to Embodiment 3 classifies a plurality of classification objects with mild cognitive impairment based on the comprehensive score (PACC (Preclinical Alzheimer Cognitive Composit)) of the clinical psychological test, namely the Preclinical Alzheimer Cognitive Function Test, performed on the classification objects, and the right hippocampal volume. Here, regarding the layering device that executes the layering method according to Embodiment 3, for the same constituent elements as those of the layering device 10, since they have already been described, the same reference numerals are given and the detailed description is omitted. The following will focus on the differences from the layering device 10.

[0185] <Configuration>

[0186] Figure 9 It is a block diagram showing an example of the configuration of the layering device 10B that executes the layering method according to Embodiment 3.

[0187] As Figure 9 shown, the difference between the layering device 10B and the layering device 10 according to Embodiment 1 is that the acquisition unit 20 is changed to the acquisition unit 20B, and the classification unit 30 is changed to the classification unit 30B.

[0188] The acquisition unit 20B acquires, for each of a plurality of classified subjects with mild cognitive impairment, a score obtained by integrating the results of several preclinical Alzheimer's cognitive function tests performed on the classified subject (hereinafter also referred to as "the integrated score of preclinical Alzheimer's cognitive function tests"), and the right hippocampal volume of the classified subject.

[0189] The classification unit 30B classifies each of the plurality of classified subjects into one of a plurality of groups including at least a first group with a relatively high risk of developing Alzheimer's disease in the future and a second group with a relatively low risk of developing Alzheimer's disease in the future, based on the integrated score of preclinical Alzheimer's cognitive function tests and the right hippocampal volume. In Embodiment 3, in addition to the first group and the second group, the plurality of groups further include a third group and a fourth group, and the plurality of groups composed of these four groups will be described.

[0190] The classification unit 30B performs the above classification based on the insights obtained by the inventors of the present invention through a cohort study on a plurality of classified subjects with mild cognitive impairment.

[0191] More specifically, the classification unit 30B performs the following classification for each of the plurality of classified subjects: (1) when the integrated score of preclinical Alzheimer's cognitive function tests is less than the third threshold and the right hippocampal volume is less than the fourth threshold, the classified subject is classified into the first group; (2) when the integrated score of preclinical Alzheimer's cognitive function tests is greater than the third threshold and the right hippocampal volume is greater than the fourth threshold, the classified subject is classified into the second group; (3) when the integrated score of preclinical Alzheimer's cognitive function tests is less than the third threshold and the right hippocampal volume is greater than the fourth threshold, the classified subject is classified into the third group; (4) when the integrated score of preclinical Alzheimer's cognitive function tests is greater than the third threshold and the right hippocampal volume is less than the fourth threshold, the classified subject is classified into the fourth group. Here, the classification unit 30B sets the fourth threshold to 0.111 and the fourth threshold to 3.651 [cm 3 , and performs the above classification.

[0192] The following describes the insights obtained by the inventors of the present invention through a cohort study on a plurality of classified subjects with mild cognitive impairment.

[0193] In a cohort study, for each of a plurality of classification objects, when starting a follow-up investigation, first, as a first-stage classification, (A) is classified into group α composed of classification objects without the apolipoprotein E ε4 gene; and (B) is classified into group β composed of classification objects with the apolipoprotein E ε4 gene. Then, the second-stage classification is performed on group α and group β classified in the first stage respectively, and the Kaplan-Meier curves showing the results of the follow-up investigation of each classified group are evaluated. The second-stage classification refers to being classified into one of the following groups (1) to (4): (1) Group A composed of classification objects with a comprehensive preclinical Alzheimer's cognitive function test score smaller than the third threshold and a right hippocampal volume smaller than the fourth threshold (corresponding to the first group in group α and the fifth group in group β); (2) Group D composed of classification objects with a comprehensive preclinical Alzheimer's cognitive function score larger than the third threshold and a right hippocampal volume larger than the fourth threshold (corresponding to the second group in group α and the sixth group in group β); (3) Group B composed of classification objects with a comprehensive preclinical Alzheimer's cognitive function score smaller than the third threshold and a right hippocampal volume larger than the fourth threshold (corresponding to the third group in group α and the seventh group in group β); and (4) Group C composed of classification objects with a comprehensive preclinical Alzheimer's cognitive function score larger than the third threshold and a right hippocampal volume smaller than the fourth threshold (corresponding to the fourth group in group α and the eighth group in group β).

[0194] Figure 10A And Figure 10B shows, in a cohort study, the Kaplan-Meier curves showing the results of the follow-up investigation for each group conducted by the inventors of the present invention.

[0195] Figure 10A is the Kaplan-Meier curve showing the results of the follow-up investigation for each group with group α as the object, Figure 10B is the Kaplan-Meier curve showing the results of the follow-up investigation for each group with group β as the object.

[0196] In Figure 10A And Figure 10B In the Kaplan-Meier curves shown in, the horizontal axis is the time [years] elapsed since the start of the follow-up investigation, and the vertical axis is the cumulative conversion rate [%] of classification objects developing from mild cognitive impairment to Alzheimer's disease.

[0197] From Figure 10A And Figure 10BAs can be seen from the Kaplan-Meier curves shown, among Groups 1 to 8, Group 1 and Group 5 are groups with a relatively high risk of developing Alzheimer's disease in the future at the start of the follow-up investigation, and Group 2 and Group 6 are groups with a relatively low risk of developing Alzheimer's disease in the future at the start of the follow-up investigation.

[0198] Thus, the inventors of the present invention obtained the following insights through a cohort study, that is, for each of a plurality of classification objects with mild cognitive impairment, by performing the following classifications (1) to (4), the plurality of classification objects with mild cognitive impairment can be classified into one of a plurality of groups (here, 4 groups) including at least Group 1 with a relatively high risk of developing Alzheimer's disease in the future and Group 2 with a relatively low risk of developing Alzheimer's disease in the future. The classifications (1) to (4) are as follows: (1) When the preclinical Alzheimer's cognitive function composite score is less than the third threshold and the right hippocampal volume is less than the fourth threshold, the classification object is classified into the first group; (2) When the preclinical Alzheimer's cognitive function test composite score is greater than the third threshold and the right hippocampal volume is greater than the fourth threshold, the classification object is classified into the second group; (3) When the preclinical Alzheimer's cognitive function test composite score is less than the third threshold and the right hippocampal volume is greater than the fourth threshold, the classification object is classified into the third group; (4) When the preclinical Alzheimer's cognitive function composite score is greater than the third threshold and the right hippocampal volume is less than the fourth threshold, the classification object is classified into the fourth group.

[0199] Furthermore, through the cohort study, the inventors of the present invention also obtained the following insights in the above classifications, that is, the first threshold is preferably 0.111 ± 1SD, and the second threshold is preferably 3.651 ± 1SD [cm 3 . Here, 1SD of the first threshold is 2.630, and 1SD of the second threshold is 0.653 [cm 3 .

[0200] <Operation>

[0201] The operation of the stratification device 10B having the above configuration will be described below with reference to the drawings.

[0202] Figure 11 is a flowchart of the third classification process performed by the stratification device 10B. This third classification process classifies each of a plurality of classification objects into one of 4 groups including at least Group 1 with a relatively high risk of developing Alzheimer's disease in the future and Group 2 with a relatively low risk of developing Alzheimer's disease in the future, based on the preclinical Alzheimer's cognitive function test composite score and the right hippocampal volume.

[0203] The third classification process is started and executed, for example, by a user using the stratification device 10B to perform an operation to start the third classification process on the stratification device 10B.

[0204] When the third classification process starts, the acquisition unit 20B acquires, for each of a plurality of classification targets with mild cognitive impairment, the score obtained from the results of the comprehensive preclinical Alzheimer's cognitive function test (PACC) performed on the classification target (hereinafter also referred to as the "PACC score"), and the right hippocampal volume of the classification target (step S205).

[0205] When the PACC scores and right hippocampal volumes of a plurality of classification targets are acquired, the classification unit 30B selects one classification target from the plurality of classification targets (step S210). Then, the classification unit 30B investigates whether the PACC score of the classification target in the selected state is 0.111 or more (step S215).

[0206] In the process of step S215, when the PACC score is not 0.111 or more (the "no" in step S215), the classification unit 30B investigates whether the right hippocampal volume of the classification target in the selected state is 3.651 or more (step S220).

[0207] In the process of step S220, when the right hippocampal volume is not 3.651 or more (the "no" in step S220), the classification unit 30B classifies the classification target in the selected state into the first group (step S225).

[0208] In the process of step S220, when the right hippocampal volume is 3.651 or more (the "yes" in step S220), the classification unit 30B classifies the classification target in the selected state into the third group (step S230).

[0209] In the process of step S215, when the PACC score is 0.111 or more (the "yes" in step S215), the classification unit 30B investigates whether the right hippocampal volume of the classification target in the selected state is 3.651 or more (step S235).

[0210] In the process of step S235, when the right hippocampal volume is not 3.651 or more (the "no" in step S235), the classification unit 30B classifies the classification target in the selected state into the fourth group (step S240).

[0211] In the process of step S235, when the right hippocampal volume is 3.651 or more (the "yes" in step S235), the classification unit 30B classifies the classification target in the selected state into the second group (step S245).

[0212] When the processing of step S225 ends, and the processing of step S230 ends, and the processing of step S240 ends, and the processing of step S245 ends, the classification unit 30B investigates whether there is an unselected classification object among the plurality of classification objects (step S250). Here, the unselected classification object refers to a classification object that is not selected in the processing of step S210, and in the loop processing from the processing of step S215 until after the "yes" processing of step S250 and then entering the processing of step S215 again, a classification object that is not yet selected in the processing of step S255 described later.

[0213] In the processing of step S250, when there is an unselected classification object (the "yes" of step S250), the classification unit 30B selects one classification object from the unselected classification objects (step S255).

[0214] When the processing of step S255 ends, the process proceeds to step S215.

[0215] In the processing of step S250, when there is no unselected classification object (the "no" of step S250), the output unit 40 outputs the classification results obtained by the classification unit 30B for each of the plurality of classification objects (step S260).

[0216] When the processing of step S260 ends, the stratification device 10B ends the third classification process.

[0217] <Consideration>

[0218] With the stratification device 10B having the above configuration, it is possible to classify a plurality of classification objects with mild cognitive impairment into one of a plurality of groups including at least a first group with a relatively high risk of developing Alzheimer's disease in the future and a second group with a relatively low risk of developing Alzheimer's disease in the future.

[0219] Therefore, by using the stratification device 10B to pre-classify a plurality of classification objects with mild cognitive impairment, it is possible to effectively conduct clinical trials and the like targeting people who develop from mild cognitive impairment to Alzheimer's disease.

[0220] Moreover, with the stratification device 10B having the above configuration, the above classification can be performed by a method different from the conventional technology. That is, in the above classification performed by the stratification device 10B, the results of brain β-amyloid PET examinations, which are relatively expensive in terms of examination costs required by the conventional technology, are not utilized.

[0221] Therefore, by using the stratification device 10B, compared with using the conventional technology, the above classification can be performed inexpensively and more effectively.

[0222] (Embodiment 4)

[0223] The following describes the layering device in Embodiment 4 that executes the layering method involved in Embodiment 4. In the configuration of Embodiment 4, a part of the functions of the layering device 10B involved in Embodiment 3 is changed. As described in Embodiment 3, the layering device 10B classifies a plurality of classification objects with mild cognitive impairment based on the comprehensive score of preclinical Alzheimer's cognitive function detection and the right hippocampal volume. Here, when classifying a plurality of classification objects with mild cognitive impairment, the layering method involved in Embodiment 4 classifies them based on whether they have the apolipoprotein E ε4 gene in addition to the comprehensive score of preclinical Alzheimer's cognitive function detection and the right hippocampal volume. Here, regarding the layering device that executes the layering method involved in Embodiment 4, since the same constituent elements as those of the layering device 10B have been described, the same numbers are given and the detailed description is omitted. The following focuses on the differences from the layering device 10B for explanation.

[0224] <Configuration>

[0225] Figure 12 It is a block diagram showing an example of the configuration of the layering device that executes the layering method 10C involved in Embodiment 4.

[0226] As Figure 12 shown, the difference in the configuration of the layering device 10C from that of the layering device 10B involved in Embodiment 3 is that the acquisition unit 20B is changed to the acquisition unit 20C, and the classification unit 30B is changed to the classification unit 30C.

[0227] For each of a plurality of classification objects with mild cognitive impairment, the acquisition unit 20C acquires, in addition to the score obtained by performing a comprehensive preclinical Alzheimer's cognitive function detection on the classification object and the right hippocampal volume of the classification object, the ApoE genotyping.

[0228] For each of the plurality of classification objects, the classification unit 30C classifies the classification object into one of a plurality of groups including at least a first group with a relatively high risk of developing Alzheimer's disease in the future and a second group with a relatively low risk of developing Alzheimer's disease in the future. In Embodiment 4, in addition to the first group and the second group, the plurality of groups also include a third group, a fourth group, a fifth group, a sixth group, a seventh group, and an eighth group. The following describes the plurality of groups composed of these 8 groups.

[0229] The classification unit 30C makes the above classification based on the insights obtained from a cohort study targeting multiple classification objects with mild cognitive impairment as the objects by the inventors of the present invention.

[0230] More specifically, the classification unit 30C classifies each of the multiple classification objects as follows (1) to (8). (1) When the ApoE genotype shows ApoE+, the comprehensive score of the preclinical Alzheimer's cognitive function test is smaller than the third threshold, and the right hippocampal volume is smaller than the fourth threshold, the classification object is classified into the first group. (2) When the ApoE genotype shows ApoE+, the comprehensive score of the preclinical Alzheimer's cognitive function test is larger than the third threshold, and the right hippocampal volume is larger than the fourth threshold, the classification object is classified into the second group. (3) When the ApoE genotype shows ApoE+, the comprehensive score of the preclinical Alzheimer's cognitive function test is smaller than the third threshold, and the right hippocampal volume is larger than the fourth threshold, the classification object is classified into the third group. (4) When the ApoE genotype shows ApoE+, the comprehensive score of the preclinical Alzheimer's cognitive function test is larger than the third threshold, and the right hippocampal volume is smaller than the fourth threshold, the classification object is classified into the fourth group. (5) When the ApoE genotype shows ApoE-, the comprehensive score of the preclinical Alzheimer's cognitive function test is smaller than the third threshold, and the right hippocampal volume is smaller than the fourth threshold, the classification object is classified into the fifth group. (6) When the ApoE genotype shows ApoE-, the comprehensive score of the preclinical Alzheimer's cognitive function test is larger than the third threshold, and the right hippocampal volume is larger than the fourth threshold, the classification object is classified into the sixth group. (7) When the ApoE genotype shows ApoE-, the comprehensive score of the preclinical Alzheimer's cognitive function test is smaller than the third threshold, and the right hippocampal volume is larger than the fourth threshold, the classification object is classified into the seventh group. (8) When the ApoE genotype shows ApoE-, the comprehensive score of the preclinical Alzheimer's cognitive function test is larger than the third threshold, and the right hippocampal volume is smaller than the fourth threshold, the classification object is classified into the eighth group. Here, the classification unit 30C sets the third threshold to 0.111 and the fourth threshold to 3.651 [cm 3 , and makes the above classification.

[0231] The insights obtained by the inventors of the present invention from a cohort study targeting multiple classification objects with mild cognitive impairment as the objects will be described below.

[0232] From Figure 10A and Figure 10B the Kaplan-Meier curves shown, among the first to eighth groups, the first group is a group with a relatively high risk of developing Alzheimer's disease in the future at the start of the follow-up survey, and the second group is a group with a relatively low risk of developing Alzheimer's disease in the future at the start of the follow-up survey.

[0233] In this way, the inventors of the present invention obtained the following insights through a cohort study, that is, by classifying each of a plurality of classified objects with mild cognitive impairment according to the following (1) to (8), it is possible to classify the plurality of classified objects with mild cognitive impairment into one of a plurality of groups (here, 8 groups) including at least a first group with a relatively high risk of developing Alzheimer's disease in the future and a second group with a relatively low risk of developing Alzheimer's disease in the future. The above classifications (1) to (8) mean: (1) When the ApoE genotyping shows ApoE+, the comprehensive score of the preclinical Alzheimer's cognitive function test is smaller than the third threshold, and the right hippocampal volume is smaller than the fourth threshold, the classified object is classified into the first group; (2) When the ApoE genotyping shows ApoE+, the comprehensive score of the preclinical Alzheimer's cognitive function test is larger than the third threshold, and the right hippocampal volume is larger than the fourth threshold, the classified object is classified into the second group; (3) When the ApoE genotyping shows ApoE+, the comprehensive score of the preclinical Alzheimer's cognitive function test is smaller than the third threshold, and the right hippocampal volume is larger than the fourth threshold, the classified object is classified into the third group; (4) When the ApoE genotyping shows ApoE+, the comprehensive score of the preclinical Alzheimer's cognitive function test is larger than the third threshold, and the right hippocampal volume is smaller than the fourth threshold, the classified object is classified into the fourth group; (5) When the ApoE genotyping shows ApoE-, the comprehensive score of the preclinical Alzheimer's cognitive function test is smaller than the third threshold, and the right hippocampal volume is smaller than the fourth threshold, the classified object is classified into the fifth group; (6) When the ApoE genotyping shows ApoE-, the comprehensive score of the preclinical Alzheimer's cognitive function test is larger than the third threshold, and the right hippocampal volume is larger than the fourth threshold, the classified object is classified into the sixth group; (7) When the ApoE genotyping shows ApoE-, the comprehensive score of the preclinical Alzheimer's cognitive function test is smaller than the third threshold, and the right hippocampal volume is larger than the fourth threshold, the classified object is classified into the seventh group; (8) When the ApoE genotyping shows ApoE-, the comprehensive score of the preclinical Alzheimer's cognitive function test is larger than the third threshold, and the right hippocampal volume is smaller than the fourth threshold, the classified object is classified into the eighth group.

[0234] And, from Figure 10A and Figure 10BAs can be seen from the Kaplan-Meier curve shown, compared with the classification result obtained by performing the above-described second-stage classification on group β composed of classification objects having the apolipoprotein E ε4 gene, the classification result obtained by performing the above-described second-stage classification on group α composed of classification objects not having the apolipoprotein E ε4 gene can more accurately classify multiple classification objects with mild cognitive impairment into one of multiple groups including at least a group with a relatively high risk of developing Alzheimer's disease in the future and a group with a relatively low risk of developing Alzheimer's disease in the future.

[0235] Thus, through the cohort study, the present inventors obtained the following insight: that is, in addition to based on the preclinical Alzheimer's cognitive function test composite score and the right hippocampal volume, further based on the ApoE genotype, the classification object can be more accurately classified into one of multiple groups including at least a first group with a relatively high risk of developing Alzheimer's disease in the future and a second group with a relatively low risk of developing Alzheimer's disease in the future.

[0236] <Work>

[0237] The operation of the stratifying device 10C having the above-described configuration will be described below with reference to the drawings.

[0238] The stratifying device 10C performs a fourth classification process, which is a process in which a part of the third classification process performed by the stratifying device 10B is changed.

[0239] Figure 13A And Figure 13B is a flowchart of the fourth classification process performed by the stratifying device 10C.

[0240] As Figure 13A And Figure 13B shown, the fourth classification process adds the process from step S300 to the process of step S380 compared with the third classification process.

[0241] Therefore, the description will be centered on the process from step S300 to the process of step S380 here.

[0242] When the fourth classification process starts, the obtaining unit 20C obtains the PACC score, the right hippocampal volume of the classification object, and the ApoE genotype for each of multiple classification objects with mild cognitive impairment (step S300).

[0243] When the process of step S300 ends, the process proceeds to the process of step S210.

[0244] When the processing of step S210 ends and the processing of step S255 ends, the classification unit 30C investigates whether the ApoE genotype of the classification target in the selected state shows ApoE+ (step S310).

[0245] In the processing of step S310, when the ApoE genotype of the classification target in the selected state does not indicate ApoE+ (the "No" of step S310), the process proceeds to the processing of step S215.

[0246] In the processing of step S310, when the ApoE genotype of the classification target in the selected state shows ApoE+ (the "Yes" of step S310), the classification unit 30C investigates whether the PACC score of the classification target in the selected state is 0.111 or more (step S320).

[0247] In the processing of step S320, when the PACC score is not 0.111 or more (the "No" of step S320), the classification unit 30C investigates whether the right hippocampal volume of the classification target in the selected state is 3.651 or more (step S330).

[0248] In the processing of step S330, when the right hippocampal volume is not 3.651 or more (the "No" of step S330), the classification unit 30C classifies the classification target in the selected state into the 5th group (step S340).

[0249] In the processing of step S330, when the right hippocampal volume is 3.651 or more (the "Yes" of step S330), the classification unit 30C classifies the classification target in the selected state into the 7th group (step S350).

[0250] In the processing of step S320, when the PACC score is 0.111 or more (the "Yes" of step S320), the classification unit 30C investigates whether the right hippocampal volume of the classification target in the selected state is 3.651 or more (step S360).

[0251] In the processing of step S360, when the right hippocampal volume is not 3.651 or more (the "No" of step S360), the classification unit 30C classifies the classification target in the selected state into the 8th group (step S370).

[0252] In the processing of step S360, when the right hippocampal volume is 3.651 or more (the "Yes" of step S360), the classification unit 30C classifies the classification target in the selected state into the 6th group (step S380).

[0253] When the processing of step S340 ends, and the processing of step S350 ends, and the processing of step S370 ends, and the processing of step S380 ends, the process proceeds to the processing of step S250.

[0254] When the processing of step S260 ends, the layering device 10C ends this fourth classification process.

[0255] <Examination>

[0256] When classifying a plurality of classification objects with mild cognitive impairment, the layering device 10C with the above configuration classifies not only according to the comprehensive score of the preclinical Alzheimer's cognitive function test and the right hippocampal volume, but also according to whether the apolipoprotein E ε4 gene is present. Accordingly, the layering device 10C can classify the classification objects more accurately.

[0257] (Embodiment 5)

[0258] The layering device in Embodiment 5 that executes the layering method according to Embodiment 5 will be described below. The configuration of Embodiment 5 is a modification of a part of the functions of the layering device 10 according to Embodiment 1. As described in the explanation of Embodiment 1, the layering device 10 classifies a plurality of classification objects with mild cognitive impairment according to the score obtained as the result of the Alzheimer's Disease Assessment Scale - Cognitive Subscale and the right hippocampal volume. Here, when classifying a plurality of classification objects with mild cognitive impairment, the layering method according to Embodiment 5 classifies in addition to the score obtained as the result of the Alzheimer's Disease Assessment Scale - Cognitive Subscale and the right hippocampal volume, also according to the score obtained from the comprehensive result of the preclinical Alzheimer's cognitive function test and whether the apolipoprotein E ε4 gene is present. Here, regarding the layering device that executes the layering method according to Embodiment 5, for the same constituent elements as the layering device 10, since they have already been described, the same numbers are given and the detailed description is omitted, and the following will focus on the differences from the layering device 10.

[0259] <Configuration>

[0260] Figure 14 It is a block diagram showing an example of the configuration of the layering device 10D in Embodiment 5 that executes the layering method according to Embodiment 5.

[0261] As Figure 14 shown, the difference in the configuration of the layering device 10D from the layering device 10 according to Embodiment 1 is that the acquisition unit 20 is changed to the acquisition unit 20D, and the classification unit 30 is changed to the classification unit 30D.

[0262] For each of a plurality of classified subjects with mild cognitive impairment, the acquisition unit 20D acquires, in addition to the score obtained as a result of the Alzheimer's Disease Assessment Scale - Cognitive Subscale for the classified subject and the right hippocampal volume of the classified subject, the score obtained as a result of the comprehensive preclinical Alzheimer's cognitive function test for the classified subject and information indicating whether the classified subject has the apolipoprotein E ε4 gene (ApoE genotyping).

[0263] For each of a plurality of classified subjects, the classification unit 30D classifies the classified subject into one of a plurality of groups including at least group 6, group 7, and group 8 with a relatively high risk of developing Alzheimer's disease in the future and group 1 and group 2 with a relatively low risk of developing Alzheimer's disease in the future, based on the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale, the right hippocampal volume, ApoE genotyping, and the comprehensive preclinical Alzheimer's cognitive function test score. In Embodiment 5, the plurality of groups include, in addition to group 1, group 2, group 6, group 7, and group 8, group 3, group 4, and group 5, and the plurality of groups composed of these 8 groups are described.

[0264] The classification unit 30D performs the above classification based on the insights obtained by the inventors of the present invention through a cohort study on a plurality of classified subjects with mild cognitive impairment as the subjects.

[0265] More specifically, the classification unit 30D classifies each of a plurality of classification targets as follows (1) to (8). (1) When the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is smaller than the fifth threshold, the volume of the right hippocampus is larger than the sixth threshold, and the ApoE genotype shows ApoE-, the classification target is classified into the first group. (2) When the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is larger than the fifth threshold, the volume of the right hippocampus is larger than the sixth threshold, the ApoE genotype shows ApoE+, and the comprehensive score of the preclinical Alzheimer's cognitive function test is larger than the seventh threshold, and when the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is smaller than the fifth threshold, the volume of the right hippocampus is larger than the sixth threshold, the ApoE genotype shows ApoE+, and the comprehensive score of the preclinical Alzheimer's cognitive function test is larger than the seventh threshold, the classification target is classified into the second group. (3) When the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is smaller than the fifth threshold, the volume of the right hippocampus is smaller than the sixth threshold, the ApoE genotype shows ApoE+, and the comprehensive score of the preclinical Alzheimer's cognitive function test is larger than the seventh threshold, the classification target is classified into the third group. (4) When the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is smaller than the fifth threshold, the volume of the right hippocampus is smaller than the sixth threshold, and the ApoE genotype shows ApoE-, the classification target is classified into the fourth group. (5) When the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is larger than the fifth threshold, the volume of the right hippocampus is larger than the sixth threshold, and the ApoE genotype shows ApoE-, the classification target is classified into the fifth group. (6) When the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is larger than the fifth threshold, the volume of the right hippocampus is larger than the sixth threshold, the ApoE genotype shows ApoE+, and the comprehensive score of the preclinical Alzheimer's cognitive function test is smaller than the seventh threshold, and when the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is smaller than the fifth threshold, the volume of the right hippocampus is larger than the sixth threshold, the ApoE genotype shows ApoE+, and the comprehensive score of the preclinical Alzheimer's cognitive function test is smaller than the seventh threshold, the classification target is classified into the sixth group. (7) When the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is larger than the fifth threshold and the volume of the right hippocampus is smaller than the sixth threshold, the classification target is classified into the seventh group. (8) When the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is smaller than the fifth threshold, the volume of the right hippocampus is smaller than the sixth threshold, the ApoE genotype shows ApoE+, and the comprehensive score of the preclinical Alzheimer's cognitive function test is smaller than the seventh threshold, the classification target is classified into the eighth group. Here, the classification unit 30D sets the fifth threshold to 20.0, the sixth threshold to 3.651 [cm 3 , and the seventh threshold to 0.111 to perform the above classification.

[0266] The following describes the insights obtained by the inventors of the present invention through a cohort study with multiple classified subjects having mild cognitive impairment as the subjects.

[0267] In the cohort study, each of the multiple classified subjects was classified into one of the following groups (1) to (8), and the Kaplan-Meier curves showing the results of the follow-up surveys of each classified group were evaluated. The groups (1) to (8) are as follows: (1) The first group consisting of classified subjects with a score of the Alzheimer's Disease Assessment Scale-Cognitive Subscale smaller than the fifth threshold, a right hippocampal volume larger than the sixth threshold, and an ApoE genotype showing ApoE-; (2) The second group consisting of classified subjects with a score of the Alzheimer's Disease Assessment Scale-Cognitive Subscale larger than the fifth threshold, a right hippocampal volume larger than the sixth threshold, an ApoE genotype showing ApoE+, and a combined score of the preclinical Alzheimer's cognitive function test larger than the seventh threshold, or classified subjects with a score of the Alzheimer's Disease Assessment Scale-Cognitive Subscale smaller than the fifth threshold, a right hippocampal volume larger than the sixth threshold, an ApoE genotype showing ApoE+, and a combined score of the preclinical Alzheimer's cognitive function test larger than the seventh threshold; (3) The third group consisting of classified subjects with a score of the Alzheimer's Disease Assessment Scale-Cognitive Subscale smaller than the fifth threshold, a right hippocampal volume smaller than the sixth threshold, an ApoE genotype showing ApoE+, and a combined score of the preclinical Alzheimer's cognitive function test larger than the seventh threshold; (4) The fourth group consisting of classified subjects with a score of the Alzheimer's Disease Assessment Scale-Cognitive Subscale smaller than the fifth threshold, a right hippocampal volume smaller than the sixth threshold, and an ApoE genotype showing ApoE-; (5) The fifth group consisting of classified subjects with a score of the Alzheimer's Disease Assessment Scale-Cognitive Subscale larger than the fifth threshold, a right hippocampal volume larger than the sixth threshold, and an ApoE genotype showing ApoE-; (6) The sixth group consisting of classified subjects with a score of the Alzheimer's Disease Assessment Scale-Cognitive Subscale larger than the fifth threshold, a right hippocampal volume larger than the sixth threshold, an ApoE genotype showing ApoE+, and a combined score of the preclinical Alzheimer's cognitive function test smaller than the seventh threshold, or classified subjects with a score of the Alzheimer's Disease Assessment Scale-Cognitive Subscale smaller than the fifth threshold, a right hippocampal volume larger than the sixth threshold, an ApoE genotype showing ApoE+, and a combined score of the preclinical Alzheimer's cognitive function test smaller than the seventh threshold; (7) The seventh group consisting of classified subjects with a score of the Alzheimer's Disease Assessment Scale-Cognitive Subscale larger than the fifth threshold and a right hippocampal volume smaller than the sixth threshold; (8) The eighth group consisting of classified subjects with a score of the Alzheimer's Disease Assessment Scale-Cognitive Subscale smaller than the fifth threshold, a right hippocampal volume smaller than the sixth threshold, an ApoE genotype showing ApoE+, and a combined score of the preclinical Alzheimer's cognitive function test smaller than the seventh threshold.

[0268] Figure 15A and Figure 15B is the Kaplan-Meier curve showing the results of the follow-up surveys for each group conducted by the inventors in the cohort study.

[0269] In Figure 15A and Figure 15B In the Kaplan-Meier curves shown, the horizontal axis is the time [years] elapsed since the start of the follow-up survey, and the vertical axis is the cumulative conversion rate [%] of the classified subjects who develop from mild cognitive impairment to Alzheimer's disease.

[0270] From Figure 15A and Figure 15B It can be seen that when the time elapsed since the start of the follow-up survey exceeds 3 years, due to the decrease in the number of classified subjects, the reliability of the cumulative conversion rate of the classified subjects who develop from mild cognitive impairment to Alzheimer's disease decreases. Therefore, in the cohort study, the value of not using the cumulative conversion rate of the classified subjects who develop from mild cognitive impairment to Alzheimer's disease when the time elapsed since the start of the follow-up survey exceeds 3 years was explored.

[0271] From Figure 15A and Figure 15B From the Kaplan-Meier curves shown, it can be known that among Groups 1 to 8, Groups 6, 7, and 8 are groups with a relatively high risk of developing Alzheimer's disease in the future at the start of the follow-up survey, while Groups 1 and 2 are groups with a relatively low risk of developing Alzheimer's disease in the future at the start of the follow-up survey.

[0272] Thus, the inventors of the present invention obtained the following insights through a cohort study. That is, by classifying each of a plurality of classified subjects with mild cognitive impairment according to the following (1) to (8), the plurality of classified subjects with mild cognitive impairment can be classified into one of a plurality of groups (here, 8 groups), including Group 6, Group 7, and Group 8 with a relatively high risk of developing Alzheimer's disease in the future, and Group 1 and Group 2 with a relatively low risk of developing Alzheimer's disease in the future. The classification of (1) to (8) means: (1) When the score on the Cognitive Subscale of the Alzheimer's Disease Assessment Scale is smaller than the fifth threshold, the volume of the right hippocampus is larger than the sixth threshold, and the ApoE genotyping shows ApoE-, the classified subject is classified into Group 1; (2) When the score on the Cognitive Subscale of the Alzheimer's Disease Assessment Scale is larger than the fifth threshold, the volume of the right hippocampus is larger than the sixth threshold, the ApoE genotyping shows ApoE+, and the comprehensive score of the Preclinical Alzheimer Cognitive Function Test is larger than the seventh threshold, and when the score on the Cognitive Subscale of the Alzheimer's Disease Assessment Scale is smaller than the fifth threshold, the volume of the right hippocampus is larger than the sixth threshold, the ApoE genotyping shows ApoE+, and the comprehensive score of the Preclinical Alzheimer Cognitive Function Test is larger than the seventh threshold, the classified subject is classified into Group 2; (3) When the score on the Cognitive Subscale of the Alzheimer's Disease Assessment Scale is smaller than the fifth threshold, the volume of the right hippocampus is smaller than the sixth threshold, the ApoE genotyping shows ApoE+, and the comprehensive score of the Preclinical Alzheimer Cognitive Function Test is larger than the seventh threshold, the classified subject is classified into Group 3; (4) When the score on the Cognitive Subscale of the Alzheimer's Disease Assessment Scale is smaller than the fifth threshold, the volume of the right hippocampus is smaller than the sixth threshold, and the ApoE genotyping shows ApoE-, the classified subject is classified into Group 4; (5) When the score on the Cognitive Subscale of the Alzheimer's Disease Assessment Scale is larger than the fifth threshold, the volume of the right hippocampus is larger than the sixth threshold, and the ApoE genotyping shows ApoE-, the classified subject is classified into Group 5; (6) When the score on the Cognitive Subscale of the Alzheimer's Disease Assessment Scale is larger than the fifth threshold, the volume of the right hippocampus is larger than the sixth threshold, the ApoE genotyping shows ApoE+, and the comprehensive score of the Preclinical Alzheimer Cognitive Function Test is smaller than the seventh threshold, and when the score on the Cognitive Subscale of the Alzheimer's Disease Assessment Scale is smaller than the fifth threshold, the volume of the right hippocampus is larger than the sixth threshold, the ApoE genotyping shows ApoE+, and the comprehensive score of the Preclinical Alzheimer Cognitive Function Test is smaller than the seventh threshold, the classified subject is classified into Group 6; (7) When the score on the Cognitive Subscale of the Alzheimer's Disease Assessment Scale is larger than the fifth threshold and the volume of the right hippocampus is smaller than the sixth threshold, the classified subject is classified into Group 7;(8) When the score on the Alzheimer's Disease Assessment Scale - Cognitive Subscale is less than the 5th threshold, the right hippocampal volume is less than the 6th threshold, the ApoE genotype shows ApoE+, and the composite score of the preclinical Alzheimer's cognitive function test is less than the 7th threshold, the classification target is classified into the 8th group.

[0273] Furthermore, the inventors of the present invention obtained the following insights through a cohort study. In the above classification, the 5th threshold is preferably 20.0 ± 1SD, the 6th threshold is preferably 3.651 ± 1SD [cm 3 , and the 7th threshold is preferably 0.111 ± 1SD. Here, 1SD of the 5th threshold is 6.3, 1SD of the 6th threshold is 0.653 [cm 3 , and 1SD of the 7th threshold is 2.630.

[0274] <Work>

[0275] Hereinafter, with reference to the drawings, the work of the hierarchical device 10D having the above configuration will be described.

[0276] Figure 16 is a flowchart of the 5th classification process performed by the hierarchical device 10D. This 5th classification process is a process of classifying each of a plurality of classification targets into one of eight groups including at least the 6th group, the 7th group, and the 8th group with a relatively high risk of developing Alzheimer's disease in the future, and the 1st group and the 2nd group with a relatively low risk of developing Alzheimer's disease in the future, based on the score on the Alzheimer's Disease Assessment Scale - Cognitive Subscale, the right hippocampal volume, the ApoE genotype, and the composite score of the preclinical Alzheimer's cognitive function test.

[0277] The 5th classification process is started, for example, by an operation of a user who uses the hierarchical device 10D to start the 5th classification process on the hierarchical device 10D.

[0278] When the 5th classification process starts, the acquisition unit 20D acquires, for each of a plurality of classification targets having mild cognitive impairment, the score obtained as a result of the Alzheimer's Disease Assessment Scale - Cognitive Subscale (ADAS - cog) for the classification target (the score of ADAS - cog), the score obtained as a result of the preclinical Alzheimer's cognitive function test comprehensive (PACC) for the classification target (the score of PACC), the right hippocampal volume of the classification target, and the ApoE genotype (step S405).

[0279] When the scores of ADAS-cog, the scores of PACC, the right hippocampal volume, and the ApoE genotype of multiple classification objects are obtained, the classification unit 30D selects one classification object from the multiple classification objects (step S410). Then, the classification unit 30D investigates whether the score of ADAS-cog of the classification object in the selected state is 20.0 or more (step S415).

[0280] In the process of step S415, when the score of ADAS-cog is not 20.0 or more ( "No" in step S415), the classification unit 30D investigates whether the right hippocampal volume of the classification object in the selected state is 3.651 or more (step S420).

[0281] In the process of step S420, when the right hippocampal volume is 3.651 or more ( "Yes" in step S420), the classification unit 30D investigates whether the ApoE genotype of the classification object in the selected state indicates ApoE+ (step S425).

[0282] In the process of step S425, when the ApoE genotype of the classification object in the selected state does not show ApoE+ ( "No" in step S425), the classification unit 30D classifies the classification object in the selected state into the first group (step S430).

[0283] In the process of step S420, when the right hippocampal volume is not 3.651 or more ( "No" in step S420), the classification unit 30D investigates whether the ApoE genotype of the classification object in the selected state indicates ApoE+ (step S435).

[0284] In the process of step S435, when the ApoE genotype of the classification object in the selected state does not show ApoE+ ( "No" in step S435), the classification unit 30D classifies the classification object in the selected state into the fourth group (step S440).

[0285] In the process of step S435, when the ApoE genotype of the classification object in the selected state shows ApoE+ ( "Yes" in step S435), the classification unit 30D investigates whether the score of PACC of the classification object in the selected state is 0.111 or more (step S445).

[0286] In the process of step S445, when the score of PACC is 0.111 or more ( "Yes" in step S445), the classification unit 30D classifies the classification object in the selected state into the third group (step S450).

[0287] In the process of step S445, when the score of PACC is not 0.111 or more (No in step S445), the classification unit 30D classifies the classification object in the selected state into the 8th group (step S455).

[0288] In the process of step S415, when the score of ADAS-cog is 20.0 or more (Yes in step S415), the classification unit 30D investigates whether the right hippocampal volume of the classification object in the selected state is 3.651 or more (step S460).

[0289] In the process of step S460, when the right hippocampal volume is 3.651 or more (Yes in step S460), the classification unit 30D investigates whether the ApoE genotyping of the classification object in the selected state indicates ApoE+ (step S465).

[0290] In the process of step S465, when the ApoE genotyping of the classification object in the selected state does not show ApoE+ (Yes in step S465), the classification unit 30D classifies the classification object in the selected state into the 5th group (step S470).

[0291] In the process of step S425, when the ApoE genotyping of the classification object in the selected state shows ApoE+ (Yes in step S425), and in the process of step S465, when the ApoE genotyping of the classification object in the selected state shows ApoE+ (Yes in step S465), the classification unit 30D investigates whether the score of PACC of the classification object in the selected state is 0.111 or more (step S475).

[0292] In the process of step S475, when the score of PACC is 0.111 or more (Yes in step S475), the classification unit 30D classifies the classification object in the selected state into the 2nd group (step S480).

[0293] In the process of step S475, when the score of PACC is not 0.111 or more (No in step S475), the classification unit 30D classifies the classification object in the selected state into the 6th group (step S485).

[0294] In the process of step S460, when the right hippocampal volume is not 3.651 or more (No in step S460), the classification unit 30D classifies the classification object in the selected state into the 7th group (step S490).

[0295] When the processing of step S430 ends, and the processing of step S440 ends, and the processing of step S450 ends, and the processing of step S455 ends, and the processing of step S470 ends, and the processing of step S480 ends, and the processing of step S485 ends, and the processing of step S490 ends, the classification unit 30D investigates whether there is an unselected classification object among the multiple classification objects (step S495). Here, the unselected classification object refers to a classification object that was not selected in the processing of step S410, and that, in the loop processing that re-enters the processing of step S415 after passing through the processing of step S415 to the "yes" processing of step S495, has not yet been selected in the processing of step S496 described later.

[0296] In the processing of step S495, when there is an unselected classification object (the "yes" of step S495), the classification unit 30D selects one classification object from the unselected classification objects (step S496).

[0297] When the processing of step S496 ends, the process proceeds to the processing of step S415.

[0298] In the processing of step S495, when there is no unselected classification object (the "no" of step S495), the output unit 40 outputs the classification results obtained by the classification unit 30D for each of the multiple classification objects (step S497).

[0299] When the processing of step S497 ends, the stratifying device 10D ends this fifth classification process.

[0300] <Consideration>

[0301] When the stratifying device 10D with the above configuration classifies multiple classification objects with mild cognitive impairment, in addition to the scores of the Alzheimer's Disease Assessment Scale - Cognitive Subscale and the right hippocampal volume, it also classifies based on the comprehensive score of the preclinical Alzheimer's cognitive function test and the presence or absence of the apolipoprotein E ε4 gene. Accordingly, the stratifying device 10D can classify the classification objects more accurately.

[0302] (Embodiment 6)

[0303] In order to more accurately obtain a stratification method for classifying each of a plurality of classification objects with mild cognitive impairment into one of a plurality of groups corresponding to the risk of developing Alzheimer's disease in the future, the inventors of the present invention further evaluated the risk of developing Alzheimer's disease from mild cognitive impairment by taking various factors as objects for positive cases of brain β-amyloid PET examination. Here, a person whose score of the result of PET examination using AV45 as a marker shows a value greater than 1.11 is referred to as a positive case of brain β-amyloid PET examination.

[0304] Figure 17 It is a schematic diagram showing an example of the content of the risk evaluation further performed by the inventors of the present invention by taking various factors as objects for positive cases of brain β-amyloid PET examination.

[0305] The inventors of the present invention Figure 17 Continuously studied and discussed the content of the risk evaluation shown, and as a result, came up with the stratification method according to Embodiment 6.

[0306] The stratification device for performing the stratification method according to Embodiment 6 will be described below. The configuration of the stratification device according to Embodiment 6 is changed in part compared with the stratification device 10 according to Embodiment 1. The stratification device according to Embodiment 6 is a device that classifies each of a plurality of classification objects with mild cognitive impairment, that is, positive cases of brain β-amyloid PET examination, into one of a plurality of groups corresponding to the risk of developing Alzheimer's disease in the future. At least a plurality of groups include a high-risk group, that is, Group 1, with a relatively high risk of developing Alzheimer's disease in the future, and a low-risk group, that is, Group 2, with a relatively low risk of developing Alzheimer's disease in the future.

[0307] The stratification device according to Embodiment 6 is an example of a device that performs classification processing, and the insights obtained by the inventors of the present invention through a cohort study taking positive cases of brain β-amyloid PET examination, that is, a plurality of classification objects with mild cognitive impairment, as objects are reflected in this classification processing.

[0308] Here, the stratification device according to Embodiment 6 has the same constituent elements as the stratification device 10. The same symbols are given to the constituent elements that have been described, and their detailed descriptions are omitted. Hereinafter, the description will be centered on the differences from the stratification device 10.

[0309] <Configuration>

[0310] Figure 18 It is a block diagram showing an example of the configuration of the stratification device 10E for performing the stratification method according to Embodiment 6.

[0311] AsFigure 18 As shown, the difference between the configuration of the stratifying device 10E and that of the stratifying device 10 according to the first embodiment is that the obtaining unit 20 is changed to the obtaining unit 20E, and the classifying unit 30 is changed to the classifying unit 30E.

[0312] For each of a plurality of classification objects who are positive for brain β-amyloid PET examination and have mild cognitive impairment, the obtaining unit 20E obtains the score obtained by the results of the Alzheimer's Disease Assessment Scale - Cognitive Subscale for the classification object, and the left hippocampal volume of the classification object. The left hippocampal volume, like the right hippocampal volume, can be measured by the following method. For example, an MRI examination is performed on the classification object, and image processing is performed on the head MRI image obtained by the results of the MRI examination for measurement.

[0313] For each of a plurality of classification objects, the classifying unit 30E classifies the classification object into one of a plurality of groups including at least a first group with a relatively high risk of developing Alzheimer's disease in the future and a second group with a relatively low risk of developing Alzheimer's disease in the future, based on the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale and the left hippocampal volume. In the sixth embodiment, in addition to the first group and the second group, the plurality of groups also include a third group and a fourth group, and the description is made with the plurality of groups being four groups.

[0314] The classifying unit 30E performs the above classification based on the insights obtained by the inventors of the present invention through a cohort study on a plurality of classification objects who are positive for brain β-amyloid PET examination and have mild cognitive impairment.

[0315] Specifically, for each of a plurality of classification objects, the classifying unit 30 performs the following classifications (1) to (4). (1) When the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is greater than the first threshold and the left hippocampal volume is smaller than the second threshold, the classification object is classified into the first group. (2) When the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is smaller than the first threshold and the left hippocampal volume is greater than the second threshold, the classification object is classified into the second group. (3) When the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is greater than the first threshold and the left hippocampal volume is greater than the second threshold, the classification object is classified into the third group. (4) When the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale is smaller than the first threshold and the left hippocampal volume is smaller than the second threshold, the classification object is classified into the fourth group. Here, the classifying unit 30E sets the first threshold to 14.0 and the second threshold to 3459 [mm 3 , and performs the above classification.

[0316] The following describes the insights obtained by the inventors of the present invention through a cohort study on multiple classified subjects who are positive for brain β-amyloid PET examination, i.e., those with mild cognitive impairment.

[0317] In the cohort study, the inventors of the present invention took multiple classified subjects who are positive for brain β-amyloid PET examination, i.e., those with mild cognitive impairment, as the objects, and tracked the development from mild cognitive impairment to Alzheimer's disease.

[0318] Figure 19 It is an example of survival prediction by the product-limit method showing the results of the follow-up investigation of each group conducted by the inventors of the present invention in the cohort study.

[0319] In Figure 19 In the survival prediction by the product-limit method shown, the horizontal axis is the time [year] elapsed since the start of the follow-up investigation, and the vertical axis is the cumulative ratio of classified subjects who do not develop from mild cognitive impairment to Alzheimer's disease.

[0320] As Figure 19 shown, in the cohort study, the inventors of the present invention took 220 classified subjects who are positive for brain β-amyloid PET examination, i.e., those with mild cognitive impairment, as the objects, and conducted a follow-up investigation for about 5 years on the development from mild cognitive impairment to Alzheimer's disease. During this period of about 5 years, 70 classified subjects developed from mild cognitive impairment to Alzheimer's disease.

[0321] In the cohort study, for each of the multiple classified subjects, at the start of the follow-up investigation, they were classified into one of the following groups (1) to (4), and the survival prediction by the product-limit method showing the follow-up results of each classified group was evaluated. The above groups (1) to (4) refer to: (1) Group A (corresponding to the first group) composed of classified subjects with a score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale greater than the first threshold and a left hippocampal volume smaller than the second threshold, (2) Group B (corresponding to the second group) composed of classified subjects with a score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale smaller than the first threshold and a left hippocampal volume greater than the second threshold, (3) Group C (corresponding to the third group) composed of classified subjects with a score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale greater than the first threshold and a left hippocampal volume greater than the second threshold, (4) Group D (corresponding to the fourth group) composed of classified subjects with a score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale smaller than the first threshold and a left hippocampal volume smaller than the second threshold.

[0322] From Figure 19As can be seen from the survival prediction by the product-limit method shown, among the first to fourth groups, the first group is a group with a relatively high risk of developing Alzheimer's disease in the future when the follow-up investigation starts, while the second group is a group with a relatively low risk of developing Alzheimer's disease in the future when the follow-up investigation starts.

[0323] Moreover, through a cohort study, the inventors of the present invention also obtained the following insights in the above classification, that is, the first threshold is preferably 14.0 ± 1SD, and the second threshold is preferably 3459 [mm 3 ± 1SD. Here, 1SD of the first threshold is 6.9, and 1SD of the second threshold is 543.5 [mm 3 .

[0324] Particularly importantly, when the first threshold and the second threshold are the above values, it can be known that the second group can be regarded as a group without the risk of developing Alzheimer's disease in the future.

[0325] <Operation>

[0326] The operation of the hierarchical device 10E having the above configuration will be described below with reference to the accompanying drawings.

[0327] Figure 20 is a flowchart of the sixth classification process performed by the hierarchical device 10E. This sixth classification process is a process of classifying each of a plurality of classification objects who are positive for the brain β-amyloid PET examination into one of four groups including at least a first group with a relatively high risk of developing Alzheimer's disease in the future and a second group with a relatively low risk of developing Alzheimer's disease in the future, according to the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale and the left hippocampal volume.

[0328] The sixth classification process is started, for example, by an operation of a user of the hierarchical device 10E to start the sixth classification process for the hierarchical device 10E.

[0329] When the sixth classification process starts, the acquisition unit 20E acquires, for each of a plurality of classification objects who are positive for the brain β-amyloid PET examination and have mild cognitive impairment, the score (score of ADAS-cog) obtained as the result of the Alzheimer's Disease Assessment Scale - Cognitive Subscale (ADAS-cog) for the classification object, and the left hippocampal volume of the classification object (step S605).

[0330] When the scores of ADAS-cog and the left hippocampal volume of multiple classification objects are obtained, the classification unit 30E selects one classification object from the multiple classification objects (step S610). Then, the classification unit 30E investigates whether the score of ADAS-cog of the classification object in the selected state is 14.0 or more (step S615).

[0331] In the process of step S615, when the score of ADAS-cog is 14.0 or more (the answer in step S615 is "yes"), the classification unit 30E investigates whether the left hippocampal volume of the classification object in the selected state is 3459 or more (step S620).

[0332] In the process of step S620, when the left hippocampal volume is not 3459 or more (the answer in step S620 is "no"), the classification unit 30E classifies the classification object in the selected state into group 1 (step S625).

[0333] In the process of step S620, when the left hippocampal volume is 3459 or more (the answer in step S620 is "yes"), the classification unit 30E classifies the classification object in the selected state into group 3 (step S630).

[0334] In the process of step S615, when the score of ADAS-cog is not 14.0 or more (the answer in step S615 is "no"), the classification unit 30E investigates whether the left hippocampal volume of the classification object in the selected state is 3459 or more (step S635).

[0335] In the process of step S635, when the left hippocampal volume is not 3459 or more (the answer in step S635 is "no"), the classification unit 30E classifies the classification object in the selected state into group 4 (step S640).

[0336] In the process of step S35, when the left hippocampal volume is 3459 or more (the answer in step S635 is "yes"), the classification unit 30E classifies the classification object in the selected state into group 2 (step S645).

[0337] When the processes of step S625, step S630, step S640, and step S645 are completed, the classification unit 30E investigates whether there is an unselected classification object among the multiple classification objects (step S650). Here, the unselected classification object refers to the classification object not selected in the process of step S610, and in the loop process that enters the "yes" process of step S650 from the process of step S615 and then enters the process of step S615 again, the classification object that has never been selected in the process of step S655 described later.

[0338] In the process of step S650, when there is an unselected classification object (Yes in step S650), the classification unit 30E selects one classification object from the unselected classification objects (step S655).

[0339] When the process of step S655 ends, the process proceeds to step S615.

[0340] In the process of step S650, when there is no unselected classification object (No in step S650), the output unit 40 outputs the classification results of each of the multiple classification objects by the classification unit 30E (step S660).

[0341] When the process of step S660 ends, the hierarchical device 10E ends the sixth classification process.

[0342] <Examination>

[0343] The hierarchical device 10E configured as described above classifies according to the scores of the Alzheimer's Disease Assessment Scale - Cognitive Subscale and the left hippocampal volume by screening out positive cases of brain β-amyloid PET examination from multiple classification objects with mild cognitive impairment. Accordingly, the hierarchical device 10E can classify classification objects with higher accuracy.

[0344] (Modification Example 1)

[0345] The hierarchical device related to Modification Example 1 will be described below. In the hierarchical device of Modification Example 1, a part of the functions of the hierarchical device 10E related to Embodiment 6 is changed. As described in Embodiment 6, the hierarchical device 10E sets the first threshold to 14.0 and the second threshold to 3459 [mm 3 , and classifies multiple classification objects with mild cognitive impairment, that is, positive cases of brain β-amyloid PET examination. In contrast, the hierarchical device related to Modification Example 1 sets the first threshold to 12.0 and the second threshold to 3737 [mm 3 , and classifies multiple classification objects with mild cognitive impairment, that is, positive cases of brain β-amyloid PET examination. That is, compared with the hierarchical device 10E, in the above classification, the hierarchical device related to Modification Example 1 changes the first threshold from 14.0 to 12.0 and the second threshold from 3459 [mm 3 to 3737 [mm 3 .

[0346] Here, for the components of the layering device related to Modification Example 1 that are the same as those of the layering device 10E, since they have already been described, the same reference numerals are given and detailed descriptions are omitted. Hereinafter, the description will focus on the differences from the layering device 10E.

[0347] <Configuration>

[0348] Figure 21 FIG. is a block diagram showing an example of the configuration of a layering device 10F that executes a layering method according to Modification Example 1.

[0349] As Figure 21 shown, the difference between the configuration of the layering device 10F and that of the layering device 10E according to Embodiment 6 is that the classification unit 30E is changed to a classification unit 30F.

[0350] Similar to the classification unit 30E, the classification unit 30F classifies each of a plurality of classification objects into one of a plurality of groups including at least a first group with a relatively high risk of developing Alzheimer's disease in the future and a second group with a relatively low risk of developing Alzheimer's disease in the future, based on the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale and the left hippocampal volume. In Modification Example 1, similar to Embodiment 6, in addition to the first group and the second group, the plurality of groups also include a third group and a fourth group. Hereinafter, the plurality of groups composed of four groups will be described.

[0351] Similar to the classification unit 30E, the classification unit 30F performs the above classification based on the insights obtained by the inventors of the present invention through a cohort study on a plurality of classification objects who are positive for brain β-amyloid PET examination and have mild cognitive impairment.

[0352] Specifically, the classification unit 30F performs the same classification as the classification unit 30E, except that the first threshold is set to 12.0 and the second threshold is set to 3737 [mm 3 .

[0353] Hereinafter, the insights obtained by the inventors of the present invention through a cohort study on a plurality of classification objects with mild cognitive impairment will be described.

[0354] In the cohort study, the inventors of the present invention took as objects a plurality of classification objects who are positive for brain β-amyloid PET examination and have mild cognitive impairment, and tracked the development from mild cognitive impairment to Alzheimer's disease.

[0355] Figure 22 FIG. shows another example of survival prediction by the product-limit method showing the results of the follow-up investigation of each group conducted by the inventors of the present invention in the cohort study.

[0356] InFigure 22 In the survival prediction by the product-limit method shown, compared with Figure 19 the survival prediction by the product-limit method shown, the horizontal axis is the time [years] elapsed since the start of the follow-up survey, and the vertical axis is the ratio of the classification objects that do not develop from mild cognitive impairment to Alzheimer's disease.

[0357] In the cohort study, for each of a plurality of classification objects, at the start of the follow-up survey, they were classified into one of the following (1) to (4) groups, and the survival prediction by the product-limit method showing the follow-up results of each classified group was evaluated. (1) Group A (corresponding to the first group) composed of classification objects with a score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale greater than the first threshold and a left hippocampal volume smaller than the second threshold. (2) Group B (corresponding to the second group) composed of classification objects with a score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale smaller than the first threshold and a left hippocampal volume greater than the second threshold. (3) Group C (corresponding to the third group) composed of classification objects with a score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale greater than the first threshold and a left hippocampal volume greater than the second threshold. (4) Group D (corresponding to the fourth group) composed of classification objects with a score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale smaller than the first threshold and a left hippocampal volume smaller than the second threshold.

[0358] From Figure 22 the survival prediction by the product-limit method shown, it can be seen that among the first to fourth groups, the first group is a group with a relatively high risk of developing Alzheimer's disease in the future at the start of the follow-up survey, and the second group is a group with a relatively low risk of developing Alzheimer's disease in the future at the start of the follow-up survey.

[0359] Furthermore, the inventors of the present invention also obtained the following insight through the cohort study, that is, in the above classification, the first threshold is preferably 12.0, and the second threshold is preferably 3737 [mm 3 .

[0360] Particularly importantly, by setting the first threshold and the second threshold to the above values, the second group can be regarded as a group without the risk of developing Alzheimer's disease in the future.

[0361] <Work>

[0362] Hereinafter, with reference to the drawings, the operation of the hierarchical device 10E configured as described above will be described.

[0363] The hierarchical device 10F performs the seventh classification process instead of the sixth classification process performed by the hierarchical device 10E.

[0364] Figure 23 is a flowchart of the seventh classification process performed by the hierarchical device 10F.

[0365] In the seventh classification process, the process from step S705 to step S710, the process from step S725 to step S730, and the process from step S740 to step S760 are respectively the same as the process from step S605 to step S610, the process from step S625 to step S630, and the process from step S640 to step S660 in the sixth classification process related to Embodiment 6, except that the classification unit 30E is replaced by the classification unit 30F and the layering device 10E is replaced by the layering device 10F. Therefore, since these processes have been described, they are omitted here. Hereinafter, the description will focus on the process of step S715, the process of step S720, and the process of step S735.

[0366] At the end of the process of step S710, the classification unit 30F investigates whether the score of ADAS-cog of the classification object in the selected state is 12.0 or more (step S715).

[0367] In the process of step S715, when the score of ADAS-cog is 12.0 or more (the "Yes" in step S715), the classification unit 30F investigates whether the left hippocampal volume of the classification object in the selected state is 3737 or more (step S720).

[0368] In the process of step S720, when the left hippocampal volume is not 3737 or more (the "No" in step S720), the process proceeds to step S725.

[0369] In the process of step S720, when the left hippocampal volume is 3737 or more (the "Yes" in step S720), the process proceeds to step S730.

[0370] In the process of step S715, when the score of ADAS-cog is not 12.0 or more (the "No" in step S715), the classification unit 30F investigates whether the left hippocampal volume of the classification object in the selected state is 3737 or more (step S735).

[0371] In the process of step S735, when the left hippocampal volume is not 3737 or more (the "No" in step S735), the classification unit 30F classifies the classification object in the selected state into the fourth group (step S740).

[0372] In the process of step S735, when the left hippocampal volume is 3737 or more (the "Yes" in step S735), the classification unit 30F classifies the classification object in the selected state into the second group (step S745).

[0373] At the end of the process in step S760, the layering device 10F ends the seventh classification process.

[0374] <Examination>

[0375] The layering device 10F configured as described above is the same as the layering device 10E. It screens out the positive cases of the brain β-amyloid PET examination from multiple classification objects with mild cognitive impairment and classifies them based on the scores of the Alzheimer's Disease Assessment Scale - Cognitive Subscale and the left hippocampal volume. Accordingly, the layering device 10F, being the same as the layering device 10E, can classify the classification objects with higher accuracy.

[0376] (Modification Example 2)

[0377] The layering device related to Modification Example 2 will be described below. In the configuration of the layering device related to Modification Example 2, a part of the functions of the layering device 10E related to Embodiment 6 is changed. As described in Embodiment 6, the layering device 10E sets the first threshold to 14.0 and the second threshold to 3459 [mm 3 and classifies multiple classification objects with mild cognitive impairment who are positive in the brain β-amyloid PET examination. Accordingly, the layering device related to Modification Example 2 sets the first threshold to 21.0 and the second threshold to 3080 [mm 3 and classifies multiple classification objects with mild cognitive impairment who are positive in the brain β-amyloid PET examination. That is, the difference between the configuration of the layering device related to Modification Example 2 and the layering device 10E is that, in the above classification, the first threshold is changed from 14.0 to 21.0, and the second threshold is changed from 3459 [mm 3 to 3080 [mm 3 .

[0378] Here, for the constituent elements of the layering device related to Modification Example 2 that are the same as those of the layering device 10E, since they have already been described, the same reference signs are given and their detailed descriptions are omitted. The following will focus on the differences from the layering device 10E for explanation.

[0379] <Configuration>

[0380] Figure 24 is a block diagram showing an example of the configuration of the layering device 10G that executes the layering method related to Modification Example 2.

[0381] As Figure 24As shown, the difference between the configuration of the stratification device 10G and the stratification device 10E according to Embodiment 6 is that the classification unit 30E is changed to the classification unit 30G.

[0382] Similar to the classification unit 30E, for each of a plurality of classification objects, the classification unit 30G classifies the classification object into one of a plurality of groups including at least a first group with a relatively high risk of developing Alzheimer's disease in the future and a second group with a relatively low risk of developing Alzheimer's disease in the future, based on the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale and the left hippocampal volume. In Modification 2, similar to Embodiment 6, in addition to the first group and the second group, the plurality of groups further include a third group and a fourth group, and the following describes these four groups.

[0383] The classification unit 30G is the same as the classification unit 30E, and performs the above classification based on the insights obtained by the inventors of the present invention through a cohort study on a plurality of classification objects who are positive for brain β-amyloid PET examination, that is, who have mild cognitive impairment.

[0384] Specifically, in the above classification performed by the classification unit 30G, except that the first threshold is set to 21.0 and the second threshold is set to 3080 [mm 3 , the rest is the same as the classification performed by the classification unit 30E.

[0385] The following describes the insights obtained by the inventors of the present invention through a cohort study on a plurality of classification objects who have mild cognitive impairment.

[0386] In the cohort study, the inventors of the present invention took a plurality of classification objects who are positive for brain β-amyloid PET examination, that is, who have mild cognitive impairment, as objects, and tracked the development from mild cognitive impairment to Alzheimer's disease.

[0387] Figure 25 Another example of survival prediction by the product-limit method showing the results of the follow-up investigation of each group performed by the inventors of the present invention in the cohort study is shown.

[0388] In Figure 25 the survival prediction by the product-limit method shown, similar to Figure 19 the survival prediction by the product-limit method shown, the horizontal axis is the time [years] elapsed since the start of the follow-up investigation, and the vertical axis is the ratio of classification objects who do not develop Alzheimer's disease from mild cognitive impairment.

[0389] In the cohort study, for each of a plurality of classification objects, at the start of the follow-up survey, they were classified into one of the following (1) to (4) groups, and the survival prediction by the product-limit method showing the follow-up results of each classified group was evaluated. The (1) to (4) groups are as follows: (1) Group A (corresponding to the first group) composed of classification objects with a score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale greater than the first threshold and a left hippocampal volume smaller than the second threshold; (2) Group B (corresponding to the second group) composed of classification objects with a score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale smaller than the first threshold and a left hippocampal volume greater than the second threshold; (3) Group C (corresponding to the third group) composed of classification objects with a score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale greater than the first threshold and a left hippocampal volume greater than the second threshold; (4) Group D (corresponding to the fourth group) composed of classification objects with a score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale smaller than the first threshold and a left hippocampal volume smaller than the second threshold.

[0390] From Figure 25 the survival prediction by the product-limit method shown, among the first to fourth groups, the first group is a group with a relatively high risk of developing Alzheimer's disease in the future at the start of the follow-up survey, and the second group is a group with a relatively low risk of developing Alzheimer's disease in the future at the start of the follow-up survey.

[0391] Moreover, the inventors of the present invention obtained the following insights through the cohort study, that is, in the above classification, the first threshold is preferably 21.0, and the second threshold is preferably 3080 [mm 3 .

[0392] Particularly importantly, by setting the first threshold and the second threshold to the above values, the first group can be regarded as a group with a very high risk that the proportion of developing Alzheimer's disease in the future is approximately 80%.

[0393] <Operation>

[0394] The operation of the hierarchical device 10G configured as described above will be described below with reference to the drawings.

[0395] The hierarchical device 10G performs the eighth classification process instead of the sixth classification process performed by the hierarchical device 10E.

[0396] Figure 26 is a flowchart of the eighth classification process performed by the hierarchical device 10G.

[0397] In the eighth classification process, the processes from step S805 to step S810, the processes from step S825 to step S830, and the processes from step S840 to step S860 respectively correspond to the processes from step S605 to step S610, the processes from step S625 to step S630, and the processes from step S640 to step S660 in the sixth classification process related to Embodiment 6. The classification unit 30E is replaced with the classification unit 30G, and the layering device 10E is replaced with the layering device 10G, and the remaining processes are the same. Here, since these processes have been described, their descriptions are omitted, and the following will focus on the processes of step S815, step S820, and step S835.

[0398] At the end of the process of step S810, the classification unit 30G checks whether the score of the ADAS-cog of the classification object in the selected state is 21.0 or more (step S815).

[0399] In the process of step S815, when the score of the ADAS-cog is 21.0 or more (the "yes" in step S815), the classification unit 30G checks whether the left hippocampal volume of the classification object in the selected state is 3080 or more (step S820).

[0400] In the process of step S820, when the left hippocampal volume is not 3080 or more (the "no" in step S820), the process proceeds to step S825.

[0401] In the process of step S820, when the left hippocampal volume is 3080 or more (the "yes" in step S820), the process proceeds to step S830.

[0402] In the process of step S815, when the score of the ADAS-cog is not 21.0 or more (the "no" in step S815), the classification unit 30G checks whether the left hippocampal volume of the classification object in the selected state is 3080 or more (step S835).

[0403] In the process of step S835, when the left hippocampal volume is not 3080 or more (the "no" in step S835), the classification unit 30G classifies the classification object in the selected state into Group 4 (step S840).

[0404] In the process of step S835, when the left hippocampal volume is 3080 or more (the "yes" in step S835), the classification unit 30G classifies the classification object in the selected state into Group 2 (step S845).

[0405] At the end of the process in step S860, the stratifying device 10G ends this eighth classification process.

[0406] <Investigation>

[0407] Similar to the stratifying device 10E, the stratifying device 10G configured as described above screens out positive cases of brain β-amyloid PET examination from multiple classification objects with mild cognitive impairment, and classifies them based on the scores of the Alzheimer's Disease Assessment Scale - Cognitive Subscale and the left hippocampal volume. Accordingly, similar to the stratifying device 10E, the stratifying device 10G can classify the classification objects with higher accuracy.

[0408] (Modification Example 3)

[0409] The stratifying device related to Modification Example 3 will be described below. In the configuration of the stratifying device related to Modification Example 3, a part of the functions of the stratifying device 10E related to Embodiment 6 has been changed. As described in Embodiment 6, the stratifying device 10E sets the first threshold to 14.0 and the second threshold to 3459 [mm 3 , and classifies multiple classification objects with mild cognitive impairment, i.e., positive cases of brain β-amyloid PET examination. In the stratifying device related to Modification Example 3, based on the scores of the Alzheimer's Disease Assessment Scale - Cognitive Subscale and the left hippocampal volume, the first threshold is set to 24.0 and the second threshold is set to 2751 [mm 3 , and classifies multiple classification objects with mild cognitive impairment, i.e., positive cases of brain β-amyloid PET examination. That is, the difference between the configuration of the stratifying device related to Modification Example 2 and the stratifying device 10E is that in the above classification, the first threshold is changed from 14.0 to 24.0, and the second threshold is changed from 3459 [mm 3 to 2751 [mm 3 .

[0410] Here, for the constituent elements of the stratifying device related to Modification Example 3 that are the same as those of the stratifying device 10E, since they have already been described, the same reference numerals are given and their detailed descriptions are omitted, and the description will be centered on the differences from the stratifying device 10E.

[0411] <Configuration>

[0412] Figure 27 is a block diagram showing an example of the configuration of the stratifying device 10H that executes the stratifying method related to Modification Example 3.

[0413] As Figure 27As shown, the difference between the configuration of the stratifying device 10H and that of the stratifying device 10E according to Embodiment 6 is that the classification unit 30E is changed to the classification unit 30H.

[0414] The classification unit 30H is the same as the classification unit 30E. For each of a plurality of classification objects, according to the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale and the left hippocampal volume, the classification object is classified into one of a plurality of groups including at least a first group with a relatively high risk of developing Alzheimer's disease in the future and a second group with a relatively low risk of developing Alzheimer's disease in the future. In Variant Example 2, the same as in Embodiment 6, in addition to the first group and the second group, the plurality of groups further include a third group and a fourth group, and these four groups will be described.

[0415] The classification unit 30H is the same as the classification unit 30E, and performs the above classification based on the insights obtained by the inventors of the present invention through a cohort study on a plurality of classification objects who are positive for brain β-amyloid PET examination, that is, those with mild cognitive impairment.

[0416] Specifically, in the above classification performed by the classification unit 30H, except that the first threshold is set to 24.0 and the second threshold is set to 2751 [mm 3 , the rest is the same as the classification performed by the classification unit 30E.

[0417] The following describes the insights obtained by the inventors of the present invention through a cohort study on a plurality of classification objects with mild cognitive impairment.

[0418] In the cohort study, the inventors of the present invention took as objects a plurality of classification objects who are positive for brain β-amyloid PET examination, that is, those with mild cognitive impairment, and tracked the development from mild cognitive impairment to Alzheimer's disease.

[0419] Figure 28 Another example of the survival prediction by the product-limit method showing the results of the follow-up survey of each group performed by the inventors of the present invention in the cohort study is shown.

[0420] In Figure 28 the survival prediction by the product-limit method shown, similar to Figure 19 the survival prediction by the product-limit method shown, the horizontal axis is the time [years] elapsed since the start of the follow-up survey, and the vertical axis is the ratio of classification objects who do not develop from mild cognitive impairment to Alzheimer's disease.

[0421] In a cohort study, for each of a plurality of classification objects, at the start of follow-up investigation, the objects were classified into one of the following groups (1) to (4), and the survival prediction by the product-limit method showing the follow-up results of each classified group was evaluated. The groups (1) to (4) are as follows: (1) Group A (corresponding to the first group) composed of classification objects with a score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale greater than the first threshold and a left hippocampal volume smaller than the second threshold; (2) Group B (corresponding to the second group) composed of classification objects with a score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale smaller than the first threshold and a left hippocampal volume greater than the second threshold; (3) Group C (corresponding to the third group) composed of classification objects with a score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale greater than the first threshold and a left hippocampal volume greater than the second threshold; and (4) Group D (corresponding to the fourth group) composed of classification objects with a score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale smaller than the first threshold and a left hippocampal volume smaller than the second threshold.

[0422] From Figure 28 the survival prediction by the product-limit method shown, among the first group to the fourth group, the first group is a group with a relatively high risk of developing Alzheimer's disease in the future at the start of follow-up investigation, while the second group is a group with a relatively low risk of developing Alzheimer's disease in the future at the start of follow-up investigation.

[0423] Moreover, the inventors of the present invention obtained the following insights through the cohort study, that is, in the above classification, the first threshold is preferably 24.0, and the second threshold is preferably 2751 [mm 3 .

[0424] Particularly importantly, by setting the first threshold and the second threshold to the above values, the first group can be regarded as a group with a very high risk, with a ratio of about 91% of developing Alzheimer's disease in the future.

[0425] <Operation>

[0426] The operation of the hierarchical device 10H configured as described above will be described below with reference to the drawings.

[0427] The hierarchical device 10H performs the ninth classification process instead of the sixth classification process performed by the hierarchical device 10E.

[0428] Figure 29 is a flowchart of the ninth classification process performed by the hierarchical device 10H.

[0429] In the ninth classification process, the processes from step S905 to step S910, the processes from step S925 to step S930, and the processes from step S940 to step S960 are respectively the same as the processes from step S605 to step S610, the processes from step S625 to step S630, and the processes from step S640 to step S660 in the sixth classification process related to Embodiment 6, except that the classification unit 30E is replaced by the classification unit 30H and the layering device 10E is replaced by the layering device 10H. Therefore, since these processes have been described, they are omitted here, and the following will mainly describe the processes of step S915, step S920, and step S935.

[0430] At the end of the process of step S910, the classification unit 30H checks whether the score of the ADAS-cog of the classification object in the selected state is 24.0 or more (step S915).

[0431] In the process of step S915, when the score of the ADAS-cog is 24.0 or more (the "yes" in step S915), the classification unit 30H checks whether the left hippocampal volume of the classification object in the selected state is 2751 or more (step S920).

[0432] In the process of step S920, when the left hippocampal volume is not 2751 or more (the "no" in step S920), the process proceeds to step S925.

[0433] In the process of step S920, when the left hippocampal volume is 2751 or more (the "yes" in step S920), the process proceeds to step S930.

[0434] In the process of step S915, when the score of the ADAS-cog is not 24.0 or more (the "no" in step S915), the classification unit 30H checks whether the left hippocampal volume of the classification object in the selected state is 2751 or more (step S935).

[0435] In the process of step S935, when the left hippocampal volume is not 2751 or more (the "no" in step S935), the classification unit 30H classifies the classification object in the selected state into Group 4 (step S940).

[0436] In the process of step S935, when the left hippocampal volume is 2751 or more (the "yes" in step S935), the classification unit 30H classifies the classification object in the selected state into Group 2 (step S945).

[0437] At the end of the process in step S960, the stratifying device 10H ends the ninth classification process.

[0438] <Examination>

[0439] The stratifying device 10H configured as described above is the same as the stratifying device 10E. It screens out positive cases of brain β-amyloid PET examination from multiple classification objects with mild cognitive impairment and classifies them based on the scores of the Alzheimer's Disease Assessment Scale - Cognitive Subscale and the left hippocampal volume. Accordingly, the stratifying device 10H can classify the classification objects with higher accuracy in the same way as the stratifying device 10E.

[0440] (Embodiment 7)

[0441] The inventors of the present invention further explored Figure 17 the content of the risk assessment shown in the figure and found a new variable. This new variable is used to classify each of multiple classification objects with mild cognitive impairment into one of multiple groups corresponding to the risk of developing Alzheimer's disease in the future.

[0442] That is, the inventors of the present invention can perform the above classification with further high accuracy by using the volume fraction value (Cog_Vol) that shows the ratio of the left hippocampal volume of the classification object to the score of the ADAS-cog of the classification object.

[0443] The example to be described below is the case where the volume fraction value (Cog_Vol) is regarded as the ratio of the left hippocampal volume [mm 3 to the score of the ADAS-cog, that is, the case where the volume fraction value (Cog_Vol) is regarded as the value obtained by dividing the left hippocampal volume [mm 3 of the classification object by the score of the ADAS-cog of the classification object.

[0444] Specifically, the volume fraction value (Cog_Vol) can be calculated by the following formula.

[0445] Cog_Vol = left hippocampal volume [mm 3 / score of the ADAS-cog

[0446] The following describes the stratification device for the stratification method involved in Embodiment 7. The stratification device involved in Embodiment 7 is constituted by changing a part of the functions of the stratification device 10E involved in Embodiment 6. Similar to the stratification device 10E, the stratification device involved in Embodiment 7 classifies each of a plurality of classification objects who are positive for brain β-amyloid PET examination, that is, those with mild cognitive impairment, into one of a plurality of groups corresponding to the risk of developing Alzheimer's disease in the future. At least among the plurality of groups, there include a high-risk group, that is, Group 1, with a relatively high risk of developing Alzheimer's disease in the future, and a low-risk group, that is, Group 2, with a relatively low risk of developing Alzheimer's disease in the future.

[0447] The stratification device involved in Embodiment 7 is an example of a device that performs classification processing using the volume fraction value (Cog_Vol) discovered by the inventors of the present invention.

[0448] Here, for the constituent elements of the stratification device involved in Embodiment 7 that are the same as those of the stratification device 10E, the same reference numerals are given as the already described constituent elements and no detailed description is made. Hereinafter, the description will be centered on the differences from the stratification device 10E.

[0449] <Constitution>

[0450] Figure 30 It is a block diagram showing an example of the constitution of the stratification device 10I that executes the stratification method involved in Embodiment 7.

[0451] As Figure 30 shown, the difference between the constitution of the stratification device 10I and that of the stratification device 10E involved in Embodiment 6 is that the classification unit 30E is changed to the classification unit 30I.

[0452] Similar to the classification unit 30E, the classification unit 30I classifies each of a plurality of classification objects into one of a plurality of groups that at least include Group 1 with a relatively high risk of developing Alzheimer's disease in the future and Group 2 with a relatively low risk of developing Alzheimer's disease in the future, based on the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale and the left hippocampal volume. In Embodiment 7, the case where the plurality of groups are composed of Group 1 and Group 2 is described.

[0453] The classification unit 30I performs the above classification based on the insights obtained by the inventors of the present invention through a cohort study of a plurality of classification objects who are positive for brain β-amyloid PET examination, that is, those with mild cognitive impairment.

[0454] Specifically, for each of a plurality of classification objects, the classification unit 30I performs the above classification based on the volume fraction value (Cog_Vol) newly discovered by the inventors of the present invention through a cohort study. Here, when the volume fraction value (Cog_Vol) is less than 101, the classification unit 30I classifies the classification object into Group 1, and when the volume fraction value (Cog_Vol) is not less than 101, the classification unit 30I classifies the classification object into Group 2.

[0455] The volume fraction value (Cog_Vol) newly discovered by the inventors of the present invention through a cohort study will be described below.

[0456] Figure 31 It is a correspondence table showing the relationship between the volume fraction value (Cog_Vol) and the number of classification objects who develop from mild cognitive impairment to Alzheimer's disease.

[0457] Figure 32 It is a correlation diagram showing the relationship, that is, showing the relationship between the volume fraction value (Cog_Vol) and the period from when the volume fraction value (Cog_Vol) is calculated until developing into Alzheimer's disease among those who are positive in the brain β-amyloid PET examination.

[0458] In Figure 32 In the shown correlation diagram, the horizontal axis is the volume fraction value (Cog_Vol), and the vertical axis is the period [year] from when the volume fraction value (Cog_Vol) is calculated until developing into Alzheimer's disease.

[0459] As Figure 31 , Figure 32 shown, among the 17 classification objects who are positive in the brain β-amyloid PET examination and have mild cognitive impairment with a volume fraction value (Cog_Vol) less than 101, 16 out of 17 developed from mild cognitive impairment to Alzheimer's disease, that is, 94.14% of the 17 classification objects with a volume fraction value (Cog_Vol) less than 101 developed from mild cognitive impairment to Alzheimer's disease.

[0460] And as Figure 32 shown, 13 out of the 17 classification objects developed from mild cognitive impairment to Alzheimer's disease within 1 year.

[0461] As described above, the inventors of the present invention discovered a new variable, the volume fraction value (Cog_Vol), which can classify a plurality of classification objects who are positive in the brain β-amyloid PET examination and have mild cognitive impairment into a group with a very high risk that the ratio of developing into Alzheimer's disease in the future is approximately 94%.

[0462] The inventors of the present invention have further explored and obtained the following insights: by setting multiple thresholds for the volume fraction value (Cog_Vol), each of multiple classification objects who are positive in the brain β-amyloid PET examination and have mild cognitive impairment can be classified in more detail, and can be classified into one of multiple groups corresponding to the risk of developing Alzheimer's disease in the future.

[0463] Figure 33 It is a correspondence table showing the relationship between the volume fraction value (Cog_Vol), multiple thresholds discovered by the inventors of the present invention, and the number of classification objects who develop from mild cognitive impairment to Alzheimer's disease, for positive cases in the brain β-amyloid PET examination among classification objects with mild cognitive impairment.

[0464] Figure 34 It is a correlation diagram showing the following relationship: showing the relationship between the volume fraction value (Cog_Vol) and the period from when the volume fraction value (Cog_Vol) is calculated until developing Alzheimer's disease, or the period elapsed without developing Alzheimer's disease, among positive cases in the brain β-amyloid PET examination.

[0465] In Figure 34 In the correlation diagram shown, the horizontal axis is the volume fraction value (Cog_Vol), and the vertical axis is the period [years] from when the volume fraction value (Cog_Vol) is calculated until developing Alzheimer's disease, or the period [years] elapsed without developing Alzheimer's disease.

[0466] Figure 35 It is a correspondence table showing the relationship between the volume fraction value (Cog_Vol), multiple thresholds discovered by the inventors of the present invention, and the trend of the ratio of classification objects who develop from mild cognitive impairment to Alzheimer's disease, for positive cases in the brain β-amyloid PET examination among classification objects with mild cognitive impairment.

[0467] As Figures 33 - 35As shown, the inventors of the present invention obtained the insight that for each of the positive cases of midbrain β-amyloid PET examination among the classified subjects with mild cognitive impairment, through the classification of (1) to (5), (1) when the volume fraction value (Cog_Vol) is less than 101 as the first threshold, the classified subject is classified into the first group; (2) when the volume fraction value (Cog_Vol) is 101 or more and 142 or less as the second threshold, the classified subject is classified into the second group; (3) when the volume fraction value (Cog_Vol) is greater than 142 as the second threshold and 266 or less as the third threshold, the classified subject is classified into the third group; (4) when the volume fraction value (Cog_Vol) is greater than 266 as the third threshold and 340 or less as the fourth threshold, the classified subject is classified into the fourth group; (5) when the volume fraction value (Cog_Vol) is greater than 340 as the fourth threshold, the classified subject is classified into the fifth group. Thus, the first group can be regarded as the group with a 94.1% ratio of developing Alzheimer's disease in the future, the second group as the group with a 60.0% ratio of developing Alzheimer's disease in the future, the third group as the group with a 32.6% ratio of developing Alzheimer's disease in the future, the fourth group as the group with a 10.4% ratio of developing Alzheimer's disease in the future, and the fifth group as the group with a 0% ratio of developing Alzheimer's disease in the future.

[0468] At this time, the inventors of the present invention obtained the following insight: the incidence rate of Alzheimer's disease within 1 year for the classified subjects in the first group is 70.5%, within 2 years is 88.2%, and within 3 years is 94.1%; the incidence rate of Alzheimer's disease within 1 year for the classified subjects in the second group is 10.0%, within 2 years is 26.6%, within 3 years is 60.0%, and within 4 years is 60.0%; the incidence rate of Alzheimer's disease within 1 year for the classified subjects in the third group is 5.9%, within 2 years is 13.9%, within 3 years is 28.7%, and within 4 years is 32.7%; the incidence rate of Alzheimer's disease within 1 year for the classified subjects in the fourth group or the fifth group is 1.4%, within 2 years is 1.4%, within 3 years is 1.4%, and within 4 years is 4.16%.

[0469] As described above, it can be seen that the greater the volume fraction value (Cog_Vol), the lower the incidence rate of Alzheimer's disease.

[0470] The inventors of the present invention conducted further investigations and obtained the following insights. That is, regardless of whether the classification object is a positive case of a brain β-amyloid PET examination, by setting multiple thresholds for the volume fraction value (Cog_Vol), each of a plurality of classification objects with mild cognitive impairment can be classified into one of a plurality of groups corresponding to the risk of developing Alzheimer's disease in the future.

[0471] Figure 36 It is a correspondence table that shows the relationship between the volume fraction value (Cog_Vol) for classification objects with mild cognitive impairment, multiple thresholds discovered by the inventors of the present invention, and the number of classification objects who develop Alzheimer's disease from mild cognitive impairment, without considering whether it is a positive case of a brain β-amyloid PET examination.

[0472] Figure 37 It is a correlation diagram that shows the relationship between the volume fraction value (Cog_Vol), the period from when the volume fraction value (Cog_Vol) is calculated until the development of Alzheimer's disease, or the period elapsed without developing Alzheimer's disease, without considering whether it is a positive case of a brain β-amyloid PET examination.

[0473] In Figure 37 the correlation diagram shown, the horizontal axis is the volume fraction value (Cog_Vol), and the vertical axis is the period [years] from when the volume fraction value (Cog_Vol) is calculated until the development of Alzheimer's disease, or the period [years] elapsed without developing Alzheimer's disease.

[0474] As Figure 36 、 Figure 37As shown, without considering whether the subject is a positive case of brain β-amyloid PET examination, the inventors of the present invention can further group each classification object with mild cognitive impairment by performing the following classifications (1) to (5). The classifications (1) to (5) are as follows: (1) When the volume fraction value (Cog_Vol) is less than the first threshold, i.e., less than 101, the classification object is classified into the first group; (2) When the volume fraction value (Cog_Vol) is greater than or equal to the first threshold, i.e., 101 or more, and less than or equal to the second threshold, i.e., 142, the classification object is classified into the second group; (3) When the volume fraction value (Cog_Vol) is greater than the second threshold, i.e., greater than 142, and less than or equal to the third threshold, i.e., 266, the classification object is classified into the third group; (4) When the volume fraction value (Cog_Vol) is greater than the third threshold, i.e., greater than 266, and less than or equal to the fourth threshold, i.e., 340, the classification object is classified into the fourth group; (5) When the volume fraction value (Cog_Vol) is greater than the fourth threshold, i.e., greater than 340, the classification object is classified into the fifth group. The grouping is as follows: The first group is regarded as the group with a 78.1% ratio of developing Alzheimer's disease in the future; the second group is regarded as the group with a 60.3% ratio of developing Alzheimer's disease in the future; the third group is regarded as the group with a 34.6% ratio of developing Alzheimer's disease in the future; the fourth group is regarded as the group with a 10.3% ratio of developing Alzheimer's disease in the future; the fifth group is regarded as the group with a 3.2% ratio of developing Alzheimer's disease in the future.

[0475] The inventors of the present invention further explored and obtained the following insight: By setting multiple thresholds for the volume fraction value (Cog_Vol) with positive cases, negative cases, and classification objects without positive or negative information of brain β-amyloid PET examination as objects, each of multiple classification objects with mild cognitive impairment can be classified into one of multiple groups corresponding to the risk of developing Alzheimer's disease in the future.

[0476] Figure 38 It is a correspondence table for positive cases, negative cases of brain β-amyloid PET examination, and classification objects without positive or negative information, that is, classification objects with mild cognitive impairment. In this correspondence table, the relationship between the volume fraction value (Cog_Vol), multiple thresholds discovered by the inventors of the present invention, and the trend of the ratio of classification objects developing from mild cognitive impairment to Alzheimer's disease is shown.

[0477] As Figure 38As shown, the inventors of the present invention obtained the following insights. That is, by classifying each of the positive subjects, negative subjects, and classification subjects with mild cognitive impairment who have no positive or negative information in the brain β-amyloid PET examination according to the following classifications (1) to (5), further grouping can be carried out. The classifications (1) to (5) are as follows: (1) When the volume fraction value (Cog_Vol) is less than the first threshold, i.e., less than 101, the classification subject is classified into the first group; (2) When the volume fraction value (Cog_Vol) is greater than the first threshold, i.e., greater than 101, and less than or equal to the second threshold, i.e., 142, the classification subject is classified into the second group; (3) When the volume fraction value (Cog_Vol) is greater than the second threshold, i.e., greater than 142, and less than or equal to the third threshold, i.e., 266, the classification subject is classified into the third group; (4) When the volume fraction value (Cog_Vol) is greater than the third threshold, i.e., greater than 266, and less than or equal to the fourth threshold, i.e., 340, the classification subject is classified into the fourth group; (5) When the volume fraction value (Cog_Vol) is greater than the fourth threshold, i.e., greater than 340, the classification subject is classified into the fifth group. The further grouping carried out is as follows: The first group is regarded as the group with a 79.4% ratio of developing into Alzheimer's disease in the future; the second group is regarded as the group with a 61.6% ratio of developing into Alzheimer's disease in the future; the third group is regarded as the group with a 36.4% ratio of developing into Alzheimer's disease in the future; the fourth group is regarded as the group with a 10.8% ratio of developing into Alzheimer's disease in the future; the fifth group is regarded as the group with a 0.6% ratio of developing into Alzheimer's disease in the future.

[0478] At this time, the inventors of the present invention obtained the following insights: the incidence rate of Alzheimer's disease within 1 year for the classification objects classified into Group 1 was 34.9%, within 2 years was 55.5%, within 3 years was 60.3%, within 4 years was 63.5%, within 5 years was 68.3%, and within 6 years was 79.4%; the incidence rate of Alzheimer's disease within 1 year for the classification objects classified into Group 2 was 15.0%, within 2 years was 30.1%, within 3 years was 45.9%, within 4 years was 53.4%, within 5 years was 54.9%, and within 6 years was 61.6%; the incidence rate of Alzheimer's disease within 1 year for the classification objects classified into Group 3 was 6.4%, within 2 years was 16.0%, within 3 years was 25.4%, within 4 years was 29.3%, within 5 years was 31.5%, and within 6 years was 36.4%; the incidence rate of Alzheimer's disease within 1 year for the classification objects classified into Group 4 was 3.2%, within 2 years was 5.4%, within 3 years was 6.5%, within 4 years was 7.6%, within 5 years was 8.7%, and within 6 years was 10.8%; the incidence rate of Alzheimer's disease within 1 year for the classification objects classified into Group 5 was 0%, within 2 years was 0%, within 3 years was 0.6%, within 4 years was 0.6%, within 5 years was 1.7%, and within 6 years was 3.4%.

[0479] As described above, the larger the volume fraction value (Cog_Vol), the lower the incidence rate of Alzheimer's disease.

[0480] The inventors of the present invention further conducted a cohort study on the classification objects among those with mild cognitive impairment for whom the results of brain β-amyloid PET examination were determined to be either positive or negative.

[0481] As a result, the inventors of the present invention obtained the following insights: by using the same as Figure 33Thresholds with different thresholds as shown can classify each of a plurality of classification objects with mild cognitive impairment, which are classification objects for which the result of a brain β-amyloid PET examination is determined to be either positive or negative, into one of a plurality of groups corresponding to the risk of developing Alzheimer's disease in the future with higher accuracy.

[0482] Figure 39 It is a correspondence table for classification objects among classification objects with mild cognitive impairment for which the result of a brain β-amyloid PET examination is determined to be either positive or negative. In this correspondence table, the volume fraction value (Cog_Vol_ab), a plurality of thresholds discovered by the inventors of the present invention, and the relationship of the transition of the ratio of classification objects developing from mild cognitive impairment to Alzheimer's disease are shown.

[0483] In Figure 39 the volume fraction value (Cog_Vol_ab) is calculated in the same way as the volume fraction value (Cog_Vol) in Figure 33 However, as described above, in addition to the two parameters of the volume fraction value (Cog_Vol) in Figure 33 this parameter of whether or not the result of the brain β-amyloid PET examination is determined is used as a third parameter, and the person for whom the result of the brain β-amyloid PET examination is determined is limited to a classification object for calculation. And Figure 39 the volume fraction value (Cog_Vol_ab) in is a value used when classifying using thresholds different from the thresholds shown in Figure 33 Therefore, in order to distinguish it from the volume fraction value (Cog_Vol) in Figure 33 it is expressed as the volume fraction value (Cog_Vol_ab).

[0484] As Figure 39As shown, the inventors of the present invention obtained the following insight: by classifying each of the classified objects for which the result of the brain β-amyloid PET examination among the classified objects with mild cognitive impairment is determined to be positive or negative as follows (1) to (5), further grouping can be performed. The classifications (1) to (5) are as follows: (1) When the volume fraction value (Cog_Vol_ab) is less than the first threshold, that is, less than 67, the classified object is classified into the first group; (2) When the volume fraction value (Cog_Vol_ab) is greater than the first threshold, that is, greater than 67 and less than or equal to the second threshold, that is, 135, the classified object is classified into the second group; (3) When the volume fraction value (Cog_Vol_ab) is greater than the second threshold, that is, greater than 135 and less than or equal to the third threshold, that is, 177, the classified object is classified into the third group; (4) When the volume fraction value (Cog_Vol_ab) is greater than the third threshold, that is, greater than 177 and less than or equal to the fourth threshold, that is, 392, the classified object is classified into the fourth group; (5) When the volume fraction value (Cog_Vol_ab) is greater than the fourth threshold, that is, greater than 392, the classified object is classified into the fifth group. The further grouping performed is as follows: the first group is regarded as the group with a 94.1% probability of developing into Alzheimer's disease in the future; the second group is regarded as the group with a 43.3% probability of developing into Alzheimer's disease in the future; the third group is regarded as the group with a 27.8% probability of developing into Alzheimer's disease in the future; the fourth group is regarded as the group with a 5.1% probability of developing into Alzheimer's disease in the future; the fifth group is regarded as the group with a 0% probability of developing into Alzheimer's disease in the future.

[0485] At this time, the inventors of the present invention obtained the following insights: the incidence of Alzheimer's disease within 1 year for the classification objects classified into Group 1 was 64.7%, within 2 years was 82.4%, within 3 years was 88.2%, and within 6 years was 94.1%; the incidence of Alzheimer's disease within 1 year for the classification objects classified into Group 2 was 11.1%, within 2 years was 13.3%, within 3 years was 34.4%, within 4 years was 38.8%, within 5 years was 38.8%, and within 6 years was 43.3%; the incidence of Alzheimer's disease within 1 year for the classification objects classified into Group 3 was 7.4%, within 2 years was 13.0%, within 3 years was 24.0%, within 4 years was 25.9%, within 5 years was 25.9%, and within 6 years was 27.8%; the incidence of Alzheimer's disease within 1 year for the classification objects classified into Group 4 was 1.3%, within 2 years was 2.5%, within 3 years was 3.2%, within 4 years was 3.2%, within 5 years was 3.8%, and within 6 years was 5.1%; the incidence of Alzheimer's disease within 6 years for the classification objects classified into Group 5 was 0%.

[0486] As can be seen from the above, the larger the volume fraction value (Cog_Vol_ab), the lower the incidence of Alzheimer's disease.

[0487] <Work>

[0488] The following describes the work of the stratifying device 10I configured as described above with reference to the drawings.

[0489] The stratifying device 10I performs the 10th classification process instead of the 6th classification process performed by the stratifying device 10E.

[0490] Figure 40 It is a flowchart of the 10th classification process performed by the stratifying device 10I.

[0491] In the 10th classification process, the processes from step S1005 to step S1010 and the processes from step S1050 to step S1060 are respectively the same as the processes from step S605 to step S610 and the processes from step S650 to step S660 in the 6th classification process related to Embodiment 6, except that the classification unit 30E is replaced with the classification unit 30I and the stratifying device 10E is replaced with the stratifying device 10I. Here, the description of the processes that have been explained will be omitted, and the following will focus on the processes from step S1015 to step S1040 for explanation.

[0492] When the process of step S1010 ends, the classification unit 30I calculates the volume fraction value (Cog_Vol) by dividing the left hippocampal volume of the classification object in the selected state by the score of the ADAS-cog of this classification object (step S1015).

[0493] Then, the classification unit 30I investigates whether the calculated volume fraction value (Cog_Vol) is less than 101 (step S1020).

[0494] In the process of step S1020, when the volume fraction value (Cog_Vol) is less than 101 (the answer is "yes" in step S1020), the classification unit 30I classifies the classification object in the selected state into Group 1 (step S1030).

[0495] In the process of step S1020, when the volume fraction value (Cog_Vol) is not less than 101 (the answer is "no" in step S1020), the classification unit 30I classifies the classification object in the selected state into Group 2 (step S1040).

[0496] When the process of step S1030 ends or the process of step S1040 ends, the process proceeds to step S1050.

[0497] When the process of step S1060 ends, the stratifying device 10I ends this 10th classification process.

[0498] <Consideration>

[0499] Similar to the stratifying device 10E, the stratifying device 10I configured as above screens out the positive cases of the brain β-amyloid PET examination from multiple classification objects with mild cognitive impairment and classifies them according to the score of the Alzheimer's Disease Assessment Scale - Cognitive Subscale and the left hippocampal volume. Accordingly, similar to the stratifying device 10E, the stratifying device 10I can classify the classification objects with higher accuracy.

[0500] (Supplement)

[0501] As described above, as an example of the technology disclosed in the present application, it has been described based on Embodiments 1 to 7 and Modification Examples 1 to 3. However, the present invention is not limited by these embodiments and modification examples. Within the scope not departing from the gist of the present invention, forms obtained by performing various modifications conceivable by those skilled in the art on this embodiment, or forms constructed by combining constituent elements in different embodiments can be included within the scope of one or more forms of the present invention.

[0502] One aspect of the present invention can be not only these layering devices 10 to 10I, but also a layering method taking the characteristic constituent parts included in the layering devices 10 to 10I as steps. Further, one aspect of the present invention can also be a computer program causing a computer to execute each of the characteristic steps included in the layering method. Moreover, one aspect of the present invention can also be a computer-readable non-transitory recording medium recording the above computer program.

[0503] In addition, the layering method according to one aspect of the present invention is effectively utilized for the preventive development of Alzheimer's disease.

[0504] Industrial Applicability

[0505] The present invention can be widely applied to a layering method or the like for classifying each of a plurality of classification objects with mild cognitive impairment into one of a plurality of groups corresponding to the risk of developing Alzheimer's disease in the future.

[0506] Symbol Explanation

[0507] 10, 10A, 10B, 10C, 10D, 10E, 10F, 10G, 10H, 10I Layering Devices

[0508] 20, 20A, 20B, 20C, 20D, 20E Acquisition Units

[0509] 30, 30A, 30B, 30C, 30D, 30E, 30F, 30G, 30H, 30I Classification Units

[0510] 40 Output Unit

Claims

1. A stratification method, the stratification method comprising: A classification step, for each of a plurality of classification objects with mild cognitive impairment, classifying the classification object into one of a plurality of groups including at least a first group with a relatively high risk of developing Alzheimer's disease in the future and a second group with a relatively low risk of developing Alzheimer's disease in the future, according to the score of the result of a clinical psychological test for the classification object and the hippocampal volume of the classification object; And An output step of outputting the classification result obtained by the classification step, The hippocampal volume of each of the plurality of classification objects is the left hippocampal volume of each of the plurality of classification objects, The clinical psychological test is the Alzheimer's Disease Assessment Scale - Cognitive Subscale, In the classification step, for each of the plurality of classification objects, a volume fraction value is calculated, the volume fraction value indicating the ratio of the left hippocampal volume of the classification object to the score of the classification object, and according to the calculated volume fraction value, the classification object is classified into one of the plurality of groups.

2. The stratification method according to claim 1, The volume fraction value is the ratio of the volume of the left hippocampus [mm 3 to the fraction. In the classification step, for each of the plurality of classification objects, when the volume fraction value is less than 101, the classification object is classified into the first group.

3. A stratification device, the stratification device comprising: An acquisition unit that acquires, for each of a plurality of classification objects with mild cognitive impairment, the score of the result of a clinical psychological test for the classification object and the hippocampal volume of the classification object; A classification unit that classifies, for each of the plurality of classification objects, the classification object into one of a plurality of groups including at least a first group with a relatively high risk of developing Alzheimer's disease in the future and a second group with a relatively low risk of developing Alzheimer's disease in the future, according to the score of the classification object and the hippocampal volume of the classification object; And An output unit that outputs the classification result obtained by the classification unit, The hippocampal volume of each of the plurality of classification objects is the left hippocampal volume of each of the plurality of classification objects, The clinical psychological test is the Alzheimer's Disease Assessment Scale - Cognitive Subscale, The classification unit calculates, for each of the plurality of classification objects, a volume fraction value, the volume fraction value indicating the ratio of the left hippocampal volume of the classification object to the score of the classification object, and according to the calculated volume fraction value, the classification object is classified into one of the plurality of groups.

4. The stratification device according to claim 3, The volume fraction value is the ratio of the left hippocampal volume [mm 3 to the fraction. The classification unit classifies, for each of the plurality of classification objects, the classification object into the first group when the volume fraction value is less than 101.

Images