Computer program and dementia functional decline likelihood determination device

Abstract

To provide a computer program and a dementia impairment occurrence possibility determination device capable of simply determining the possibility of the occurrence of a mild cognitive impairment (MCI).SOLUTION: A computer program causes a computer having a screen and voice output means to execute: a display step of displaying a predetermined problem on the screen; an answer step of instructing a subject to answer, with voice by the voice output means; a storage step of acquiring a result of making the subject to answer, determining whether the answer is correct, and stores a result of the determination in a recording medium; and a determination step of determining the possibility of the occurrence mild cognitive impairment in the future by substituting the stored result into a predetermined expression and comparing a result of the substitution with a predetermined threshold.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a program and a device for determining the likelihood of dementia functional decline, which can easily determine the likelihood of mild cognitive impairment (MCI) occurring. [Background technology]

[0002] In the current situation where there are no established therapeutic or preventive medications for dementia, the most effective measure is to detect and treat mild cognitive impairment (MCI), a precursor to dementia, early on. The most commonly used method for assessing cognitive function is paper-based psychological testing, which requires trained examiners and has problems such as a learning effect due to repeated testing, making it difficult for anyone to easily assess cognitive function at any time. Therefore, various methods for easily evaluating cognitive function and measures against dementia have been proposed. For example, Patent Document 1 proposes to provide a dementia prevention system, a dementia prevention method, and a program that can promote improvement of a user's brain function related to long-term memory and improvement of lifestyle habits. Specifically, the proposed system includes a problem storage unit that stores a plurality of training problems classified according to difficulty level and lifestyle habit advice, a user information storage unit that stores a plurality of user information items that associate a user identifier with difficulty level information of the problems, a selection unit that selects a training problem as an output problem from the plurality of training problems based on the difficulty level information of the problems associated with the user information of the user when a training start request is received from a terminal device, a creation unit that creates a past confirmation question related to at least one of the output problem and the lifestyle habit advice output to the user, and a system that outputs at least one of the output problem selected by the selection unit, the past confirmation question created by the creation unit, and the lifestyle habit advice stored in the problem storage unit. Furthermore, Patent Document 2 proposes to provide a cognitive function assessment device, a cognitive function assessment method, and a program that can obtain highly accurate cognitive function assessment results in a short time. Specifically, in the cognitive function assessment device, a formant analysis unit receives target data representing time fluctuations in the instantaneous sound pressure of specific phonemes contained in the subject's voice over a target period. The formant analysis unit then divides the target period into multiple frames and determines the frequency of a specific formant for each of two or more target frames. A feature analysis unit determines a feature for the specific formant frequency determined for each target frame. The proposed device includes an evaluation unit that evaluates the subject's cognitive function based on the feature. Furthermore, Patent Document 3 proposes a system that enables users, medical professionals, teachers, and parents to target, individually measure, and manage cognitive skill development. Specifically, as a game-based virtual learning curriculum for targeting and developing cognitive skills underlying executive function, an effective and rapid video game-based training curriculum for improving cognitive skills (e.g., focused attention, sustained attention, cognitive inhibition, behavioral inhibition, selective attention, shifting attention, allocated attention, interference control, novelty inhibition, delayed gratification tolerance, inner voice, motivational inhibition, and self-control) is proposed. Furthermore, computer-based cognitive function assessment methods have been reported in Non-Patent Documents 1 and 2, and Non-Patent Document 3 reports on computer-based prediction of dementia onset. Furthermore, as shown in Non-Patent Document 4, the present inventors have already reported on the possibility of cognitive function testing using tablet devices. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2019-219538 [Patent Document 2] Japanese Patent Application Publication No. 2018-050847 [Patent Document 3] Special Publication No. 2018-533044 [Non-patent literature]

[0004] [Non-Patent Document 1] Makizako H, Shimada H, Park H, Doi T, Yoshida D, Uemura K, et al. Evaluation of multidimensional neurocognitive function using a tablet personal computer: test-retest reliability and validity in community-dwelling older adults. Geriatr Gerontol Int. 2013;13(4):860-6. [Non-patent document 2] Paul Maruff YYL, David Darby, Kathryn A Ellis, Robert H Pietrzak, Peter J Snyder, Ashley I Bush, Cassandra Szoeke, Adrian Schembri, David Ames, Colin L Masters. Clinical utility of the cogstate brief battery in identifying cognitive impairment in mild cognitive impairment and Alzheimer's disease. BMC Pharmacology and Toxicology. 2013;1. [Non-patent document 3] Shimada H, Makizako H, Park H, Doi T, Lee S. Validity of the National Center for Geriatrics and Gerontology-Functional Assessment Tool and Mini-Mental State Examination for detecting the incidence of dementia in older Japanese adults. Geriatr Gerontol Int. 2017;17(12):2383-8. [Non-patent document 4] Abstracts of the 77th Annual Meeting of the Japanese Association of Public Health, published October 9, 2018, presentation number O-2301-5, title: "Development of a cognitive function test using a tablet computer and its validity" Summary of the Invention [Problem to be solved by the invention]

[0005] However, the devices and systems proposed in the above-mentioned patent documents and non-patent documents have not yet suggested any way to predict mild cognitive impairment (MCI). If it were possible to easily and objectively predict the onset of MCI, various measures could be taken to prevent progression to dementia. Furthermore, those affected would be able to recognize the risk at an early stage, giving them time to prepare for living with cognitive decline. Furthermore, the report in Non-Patent Document 4 still had the problem of not being able to perform cognitive function tests with sufficient accuracy. Therefore, an object of the present invention is to provide a computer program and a device for determining the possibility of dementia functional decline, which can easily and accurately determine the possibility of the occurrence of mild cognitive impairment (MCI). [Means for solving the problem]

[0006] As a result of intensive research into solving the above problems, the inventors discovered that the above object can be achieved by carefully selecting problems and performing ROC analysis using the results obtained from these problems, and have completed the present invention based on this discovery. That is, the present invention provides the following inventions. 1. A computer program that causes a computer having a screen and audio output means to execute the following steps: a display step of displaying a predetermined question on the screen; an answering step of giving voice instructions to the subject by the voice output means to have the subject answer; a storage step of acquiring the answers given by the subject, determining whether the answers are correct or incorrect, storing the score, age, and gender information on a recording medium, and substituting the score, age, and gender information into a predetermined formula to obtain a predicted score; A determination step in which the stored predicted score is compared with a predetermined threshold to determine the likelihood of future occurrence of mild cognitive decline. 2. The audio output means is a noise-canceling headphone. 1. A computer program according to claim 1. 3. In the display step, the problems displayed on the screen are number recitation, number recitation backward, immediate memory, remote memory, Stroop task, and figure recognition; 3. The computer program according to claim 1 or 2, wherein the predetermined formula in the storage step is the following prediction formula for the occurrence of mild cognitive decline: Prediction formula for mild cognitive decline Predicted score = Number recitation score × (-0.503) + Number recitation score × (-0.540) + Immediate memory score × (-0.056) + Remote memory score × (-0.314) + Stroop task score × (-0.155) + Shape recognition score × (-0.307) + Gender (Male: -0.233, Female: 0) + Age × (0.00036) + Constant (-3.105) 4. A computer program according to any one of 1 to 3, characterized in that in the storage step, when obtaining the predicted score using the predetermined formula, the response time is also obtained and the judgment result is corrected by weighting it with the response time. 5. A dementia function decline likelihood determination device storing the computer program described in 1, having a screen and an audio output means, a display means for displaying a predetermined question on the screen; answering means for giving voice instructions to the subject by the voice output means to cause the subject to answer; a storage means for acquiring the answers given by the subject, determining whether the answers are correct or incorrect, and storing the determination result in a recording medium; A dementia function decline likelihood determination device comprising a determination means step of substituting the stored result into a predetermined formula, comparing the substitution result with a predetermined threshold, and determining the likelihood of future mild cognitive decline. [Effects of the Invention]

[0007] The computer program of the present invention can propose and provide a device for determining the possibility of dementia functional decline, which can easily determine the possibility of mild cognitive impairment (MCI) occurring, and the device for determining the possibility of dementia functional decline of the present invention can easily determine the possibility of mild cognitive impairment (MCI) occurring. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a schematic diagram of a computer used in the present invention. [Figure 2] FIG. 2 is a schematic diagram showing a flow sheet of the computer program of the present invention. [Figure 3] 3(a) to 3(f) are explanatory diagrams showing examples of questions displayed on a computer screen using the computer program of the embodiment shown in FIG. [Figure 4] FIG. 4 is an explanatory diagram illustrating another problem. [Figure 5] FIG. 5 is a chart showing the ROC curve in Example 1. [Figure 6] FIG. 6 is a chart showing the ROC curve in Example 2. [Explanation of symbols]

[0009] 1: Computer, 11: Memory, 13: CPU, 15: Storage medium, 20: Input device, 30: Screen, 40: Audio output means DETAILED DESCRIPTION OF THE INVENTION

[0010] The present invention will now be described in further detail. First, the computer program of the present invention (hereinafter, sometimes simply referred to as the "program") will be described, followed by the device for determining the likelihood of dementia functional decline of the present invention (hereinafter, abbreviated as the "determination device"). The program of this embodiment is for conducting a predetermined test on a subject and determining the likelihood of dementia functional decline in the future based on the results. The computer program of the present invention causes a computer having the screen and audio output means shown in FIG. 1 to execute the following steps shown in FIG. a display step S1 for displaying a predetermined question on the screen; an answering step S2 in which the subject is instructed by voice using the voice output means to answer; A storage step S3 in which the answers given by the subject are acquired, the correctness or incorrectness is determined, and the score, age, and gender information are stored on a recording medium, and the score, age, and gender information are substituted into a predetermined formula to obtain a predicted score. A determination step S4 compares the stored predicted score with a predetermined threshold to determine the likelihood of future occurrence of mild cognitive decline. In the following, the computer will be described first, and then each step of this embodiment will be described.

[0011] 〔computer〕 Specifically, the computer 1 used in this embodiment can be any ordinary personal computer including a central processing unit (CPU) 13, memory 11 as a temporary storage area, and a non-volatile storage medium 15 such as a hard disk or solid-state device, as shown in Fig. 1. Furthermore, although not specifically shown, the computer in this embodiment preferably has a communications device and is capable of communication via a network. It can also be configured to connect to a server having a database placed on the network through communications and obtain updated data from the database as needed. The computer is also equipped with input devices 20, such as an image input device such as a keyboard, mouse, or camera, an audio input device such as a microphone, and a communication input device using a communication device such as Bluetooth (registered trademark), and is configured to input necessary data and information as needed. It is also equipped with a screen (display) 30 for displaying the evaluation results, and the results are displayed in a desired format as needed. In this embodiment, the screen 30 is a touch panel with a built-in touch sensor, and answers and operations described below can be entered by directly touching the screen. Furthermore, by connecting the computer to a printer (not shown), necessary information, such as the evaluation results, can be output as needed. In this embodiment, the computer is provided with audio output means 40, specifically external output means such as a speaker or headphones. The audio output means is preferably so-called noise-canceling headphones. That is, the audio output means 40 may be built into the computer, or may be external output means such as headphones connected via the computer's earphone jack or via a communication device such as Bluetooth (registered trademark). In this embodiment, the program of this embodiment is stored in this computer, and the computer functions as a display means, a response means, a storage means, and a determination means, which will be described later, and functions as a determination device. In addition, in this embodiment, it is preferable to use a portable terminal such as a smartphone or tablet terminal, whose screen also serves as an input means, and these are also included in the "computer" of the present invention. <Other components (devices)> The computer of this embodiment may include various devices other than the above-mentioned devices as necessary.

[0012] [Preliminary step S0] The computer program of this embodiment preferably includes a preliminary step of creating a database containing questions (question data displayed on the screen and audio questions output in response to the question data) and answer data by storing predetermined questions for conducting predetermined tests together with their answers in the computer or a server connected to the computer via a network. This database also stores subject information such as gender, age, medical history, and years of education, as well as the percentage of correct answers to the questions. It also stores scores obtained when multiple subjects administer the test stored in the program of this embodiment, and data on whether or not the subject has previously experienced cognitive decline and how many years (or months) since the test that this occurred. Furthermore, as described below, a threshold is set based on the data stored in the database.

[0013] [Display step S1] The display step S1 is a step for displaying predetermined questions on the screen in order to conduct a predetermined test. Note that the display step S1 and the answer step S2 may not be completely separated depending on the question, and the display step S1 and the answer step S2 may be performed intermixed. Therefore, in the explanation of this step, the answer step S2 may also be explained as appropriate. The questions displayed on the screen here include a multi-domain cognitive function test, with specific question categories including: 1. reciting numbers forward, 2. reciting numbers backward, 3. immediate memory, 4. remote memory, 5. Stroop task, and 6. figure recognition. These tests have criterion-related validity with the dementia screening test (MMSE), with a correlation coefficient of 0.51. These are examples, and other questions may be added or the example questions may be deleted. The screen display, the purpose of the questions, and how to answer each question are explained below. In the number recitation, one number from 0 to 9 is randomly selected and displayed in the center of the screen. After a certain period of time (approximately one second), the displayed number disappears from the screen. This process is repeated, and initially, a three-digit number (a three-digit number is indicated by displaying a number three times) is displayed on the screen. In the answer step S2 described below, the numbers 0 to 9 are displayed in a single row of a table at the bottom of the screen, and three blank cells are displayed in a single row of a table in the middle of the screen (the number of blank cells increases according to the number of digits in the number asked). The respondent answers by selecting (tapping) the numbers in the order they were initially displayed. The selected number is displayed in the middle cell. As the respondent answers correctly, the number of digits displayed on the screen increases from three to five, five to seven, and finally eight digits are displayed (in Figure 3(a) , the middle cell shows the first three numbers being answered for the five numbers read out). As shown in Figure 3(b), the screen configuration and display of numbers in the question for reciting numbers backwards are the same as for reciting numbers forwards. That is, in reciting numbers backwards, the display step S1 is the same as for reciting numbers forwards. However, when answering (answer step S2), the numbers displayed on the screen are selected (tapped) in the reverse order of their display. In the immediate memory test, 10 words are randomly displayed on the screen. The words are displayed one by one on the screen for memorization (not shown). The display time for each word is preferably about 1 second (this is the display step S1). Then, in the answer step S2, as shown in FIG. 3(c), a total of 20 words are displayed on the screen, including 10 dummy words and the 10 words actually displayed on the screen. In this embodiment, the subject answers by tapping on one of the 10 words displayed for memorization. At this time, the tapped word is set to change color (in this embodiment, it changes from white to green). The subject is instructed to remember the words displayed by voice, as the words displayed in the immediate memory test will be asked to answer again during the remote memory test, which is the final item of the test. In the Stroop task, in the display step S1, as shown in Figure 3(d), a word indicating a color is displayed in a color different from the color that the word represents (in this embodiment, the word "red" is displayed in large letters in the center in blue), and the word indicating the color is displayed in the table below. Then, in the answer step S2, the subject answers by selecting / tapping the color of the letter displayed in the center from the table below in accordance with audio instructions. In the figure recognition, in the display step S1, a figure formed by arranging multiple cubes is displayed, and numbers are displayed in a table below, as shown in Fig. 3(e). Then, in the answer step S2, the test subject answers by selecting and tapping the number of cubes that make up the figure from the numbers shown in the table below. In the display step S1, the remote memory displays words in a table format as shown in Fig. 3(f). Then, in the answer step S2, the user answers by selecting and tapping the word displayed in the immediate memory from the words displayed in the table format.

[0014] [Answer step S2] This step is a step in which the subject is instructed by voice using the voice output means, and an answer field is displayed as necessary, to have the subject answer. Also, as explained in the above section on display step S1, the screen display in display step S1 may differ from the screen display in answer step S2, and in that case, the screen display is performed as described above in answer step S2. That is, in answer step S2, not only voice output but also display on the screen may be performed. In this embodiment, the audio output means is a so-called noise-canceling headphone. Here, noise-canceling headphones are headphones capable of reducing ambient noise and providing clear sound. They typically have a built-in microphone that collects ambient noise, detects the frequency of the noise, and generates a sound with an opposite phase to the detected frequency, thereby canceling the noise. Note that the term "headphones" also includes so-called earphone-type headphones, such as so-called in-ear headphones. Commercially available noise-canceling headphones can be used without any particular restrictions. In this step, the timing and content of execution differ depending on the category of the question displayed in the display step S1 described above. Below, the content set to execute this step corresponding to the question category will be explained. In this step, a question voice corresponding to the displayed question is output from the computer by audio output. At this time, the question voice is randomly selected from the question voices stored in the database and output, completing the question. Once the question is completed, the subject will answer. Completion of question setting will be explained for each type of question. When reciting the numbers in order, the question is completed by outputting the instruction text for the question (in this embodiment, "Next, select and touch the numbers that will be displayed on the screen in the order they appear from the column below"). When reciting numbers backwards, the question is completed by outputting the instruction text for the question (in this embodiment, "Next, select and touch the numbers that will be displayed on the screen in the reverse order from the column below") aloud. In instant memorization, after all the words displayed sequentially for memorization are output, the instruction text for the question (in this embodiment, "Please touch all the words that were previously displayed on the screen in the table that will be displayed next") is output aloud to complete the question. In the Stroop task, the question is completed by outputting the instruction text of the question (in this embodiment, "Please select the color of the next character to be displayed from the box below and touch it") aloud. In the shape recognition, the question is completed by outputting the instruction text of the problem (in this embodiment, "Next, select the number of cubes used in the shape from the box below and touch it") aloud. In remote memory, the question is completed by outputting the instruction text of the question (in this embodiment, "Please select and touch the subword that was previously displayed on the screen from the column below"). In this embodiment, the questions are displayed and the answers are stored (a storage step described later), and a step of acquiring the subject's age and gender is also performed. This step can be performed in the answering step by displaying a screen for inputting subject information and having the subject enter their age and gender prior to answering the questions. In addition, since the questions that have been asked need to be checked for authenticity in the storage step, the questions that have been asked are stored in a storage medium so that it is possible to reconfirm which question has been asked and in which category. Then, when the questions have been presented as described above, the subject answers by tapping the numbers or words displayed on the screen corresponding to each of the questions, in response to the screen displayed and the audio question output. If the subject confirms the answer, or since a predetermined answer time is set for each type of question, in that case, even if the answer has not been given after the answer time has elapsed, the answer is considered complete, and the process proceeds to storage step S3.

[0015] [Storage step S3] This step involves obtaining the answers given by the subject, determining whether they are correct or incorrect, calculating the score, storing the score, age, and gender on a recording medium, and substituting the score, age, and gender information into a specified formula to obtain a predicted score. First, the answer results are acquired. This acquisition is performed for each question. In other words, if multiple questions are asked in the same category, the answer results are acquired for each question. Specifically, in the case of reciting numbers in order, the numbers are read out, the subject taps the answer, and when the subject completes the answer up to the final question (or the set answer time has elapsed), the answered number is obtained. The answer time (the time from the audio question output to touching) is also obtained. In this case, the obtained data can be saved in the computer's memory and then stored in the computer's recording medium, or it can be stored in the computer's recording medium each time the subject taps. Reciting numbers backward is the same as reciting numbers in order. In immediate memory, when the subject taps all words (all 10 words if 10 words are read out) and completes the answer (or the set answer time has elapsed), the answered word is obtained. The answer time (the time from the audio question output to the completion of touching) is also obtained. In this case, the obtained data is stored in the computer's recording medium. In the Stroop task, when the subject taps a word and completes the answer (or the set answer time has elapsed), the answered word is obtained. The answer time (the time from the audio question output to the completion of touching) is also obtained. In this case, the obtained data is stored in the computer's recording medium. With graphic recognition, the subject taps a number and the answer is completed (or the set answer time has passed), and the answering word is acquired. The answering time (the time from the audio output of the question to the completion of touching) is also acquired. At this time, the acquired data is stored in the computer's recording medium. With remote memory, the subject taps all the words (all 10 if 10 words are read out) and the answer is completed (or the set answer time has passed), and the answering word is acquired. Next, the acquired answer is compared with the answer to the question that has been temporarily stored in the storage medium to determine whether the answer of the subject is correct or not. The test subject's score is then calculated based on the results of the judgment, and the correct or incorrect answer for each question and the score are stored in a storage medium. The obtained score is then substituted into the following prediction formula for the occurrence of mild cognitive decline, which has been created in advance. The subject's age and gender, acquired in response step S2, are also substituted into the prediction formula below. Specifically, the obtained score and information on age and gender are substituted to calculate a predicted score for mild cognitive decline. The obtained predicted score is then stored in a recording medium. Prediction formula for mild cognitive decline Predicted score = Number recitation score × (-0.503) + Number recitation score × (-0.540) + Immediate memory score × (-0.056) + Remote memory score × (-0.314) + Stroop task score × (-0.155) + Shape recognition score × (-0.307) + Gender (Male: -0.233, Female: 0) + Age × (0.00036) + Constant (-3.105) Note: For gender, enter -0.233 for males and 0 for females. The display step S1, answer step S2, and storage step S3 are all performed for each question. In this embodiment, the display step S1 and answer step S2 are looped for each question type, and the storage step S3 is set to store the results based on the answer data obtained in the answer step S2.

[0016] [Decision step S4] This step compares the stored predicted score with a predetermined threshold to determine the likelihood of future mild cognitive decline, where "future" means within two years of the subject answering the above questions (taking the test). The above-mentioned predetermined threshold is determined by obtaining an ROC curve for the onset of dementia function decline within two years based on the data in the above-mentioned database, and calculating the threshold based on the obtained ROC curve using a standard method (e.g., Youden's index) (see the Examples for specific values). This is the threshold for the prediction score in the database. Note that the threshold for the prediction score in the following Examples is 0.0772. When making the judgment, first the predicted score of the test subject is calculated. That is, when all questions have been answered, all correct / incorrect answers, age, and gender (age and gender can be stored at any stage of the storage step, and in some cases can also be stored in the display step or pre-step) are stored, and these stored results (all correct / incorrect answers, age, and gender) are substituted into a predetermined prediction formula to calculate the predicted score. The judgment is made by comparing the subject's predicted score obtained here with the predicted score as the threshold value mentioned above, and if the subject's predicted score is below the threshold value, it is judged that ``there is a high possibility that cognitive function will decline in the near future (within two years).'' The results of the determination in this step are stored in a storage medium in association with personal information such as the subject's name, age, and sex, and are also displayed on a computer screen.

[0017] [Correction of Storage Step S3 and Determination Step S4] In addition, in this step, the answers given by the subject are obtained, judged as correct or incorrect, the answer time is also obtained along with the judgment result, the judgment result is weighted and corrected by the answer time, and the obtained score can be stored on a recording medium. That is, since the subject's response time is also acquired in the response step S2, the storage step S3 determines whether the response time is fast or slow, i.e., compares it with the average response time and weights the score depending on whether it is faster than the average, the same as the average, or slower than the average. In this case, the score is corrected by multiplying it by a predetermined coefficient as a weighting factor for the response time. The corrected score is then stored in a recording medium, and the above-mentioned predicted score can be calculated using this corrected score. Here, the predetermined coefficient can be, for example, a weighting coefficient of 3 if faster than the average, 2 if the same as the average (within a range of ±1), or 1 if slower than the average.

[0018] (Other steps) The computer program of this embodiment may perform other steps without departing from the spirit of the present invention. In addition to the above problems, the computer program of this embodiment may also be configured to execute the following additional problems. Additional questions: Trail making test (TMT) As shown in Figure 4, the hiragana characters "A, I, U, E, O, Ka, Ki, Ku" and the numbers "1, 2, 3, 4, 5, 6, 7" are displayed randomly across the entire screen. The subject selects the hiragana characters and numbers alternately in order. The selected letters and numbers are displayed on the screen so that they are connected by lines. For example, in the example shown in Figure 4, if the subjects select "A → 1 → I → 2 → U → 3 → E → 4 → O," each hiragana character and number will be connected by a line, as shown in Figure 4. 〔Device〕 Next, a dementia function decline possibility determination device in which the computer program of this embodiment is stored will be described. The dementia function decline possibility determination device of this embodiment is configured by a computer shown in FIG. That is, it is configured by a computer 1 in which the above-mentioned computer program is stored, and has a screen 30 and audio output means 40. The CPU 13, memory 11, storage medium 15, and screen 30 constitute a display means for displaying predetermined questions stored in the storage medium 15 on the screen 30, the CPU 13, memory 11, storage medium 15, screen 30, and audio output means 40 constitute an answering means for instructing the subject by voice to answer, the CPU 13, memory 11, storage medium 15, and screen 30 constitute a storage means for acquiring the answer results given by the subject, determining whether they are correct or incorrect, and storing the determination result in the recording medium 15, and the CPU 13, memory 11, storage medium 15, and screen 30 constitute a determination means for substituting the stored result into the above-mentioned formula stored in the storage medium, comparing the substitution result with the above-mentioned determination criterion, which is a predetermined threshold, to determine the possibility of mild cognitive decline occurring in the future (within 2 years).

[0019] [Usage example] The dementia function decline possibility determination device storing the computer program of this embodiment is used as follows. That is, an icon (not shown) (corresponding to the start button of the computer program of this embodiment, such an icon can be set or can be set as a function of the computer program of this embodiment) is displayed on the computer screen. When the subject taps this icon, the computer program is executed, and a start screen for the questions (not shown) is displayed. This start screen displays an input window for inputting the subject's personal information and a start button. By inputting the personal information in the input window and pressing the start button, the display step S1 begins. Then, as shown in FIG. 2, the process proceeds to an answer step S2 for each question, and when the answer is completed, the process proceeds to a storage step S3. These steps S1 to S3 are executed until all questions are completed. Then, when all questions are completed, the process proceeds to the next judgment step S4. The process can be set to automatically proceed to the judgment step S4 when all questions are completed, or a judgment button may be displayed when the questions are finished, and a judgment is made when the judgment button is tapped. In this way, the determination step S4 is performed, and the determination result is displayed on the screen (not shown).

[0020] [Action and effect] The computer program and dementia functional decline likelihood determination device of this embodiment are configured as described above and can be used as described above. The subject can answer by looking at the questions displayed on the screen and listening to the audio output, and even elderly people can answer smoothly without the help of an assistant.The possibility of future (within two years) dementia decline can be easily determined by simply answering a few questions using a computer with a touch panel as the screen. Furthermore, when noise-canceling headphones are used, the subject's malfunctions are particularly reduced, and accurate response data (correct / incorrect and response time) can be obtained, allowing for a more accurate assessment of the possibility of dementia functional decline. For these reasons, the computer program and dementia function decline likelihood determination device of this embodiment have the advantage of enabling cognitive function testing without the need for trained staff at community health checkups, events for the elderly, etc. In addition, cognitive function can be evaluated using a PC or smartphone at home without having to go to a medical institution. Therefore, it is expected to be used in situations where a simple estimation of cognitive function is required, such as in medical settings and local health and welfare services.

[0021] The present invention is not limited to the above-described embodiment, and various modifications are possible without departing from the spirit of the present invention. For example, the database can be set to be updated sequentially. In this case, the threshold value is updated as the amount of data stored in the database increases, making it possible to make a judgment using a threshold value that allows for more accurate judgment. [Example]

[0022] The present invention will be specifically explained below with reference to examples, but the present invention is not limited to these examples. [Example 1] (Example of confirmation of the accuracy of a judgment using the program of the present invention) The criterion-related validity of the dementia function decline assessment device shown in Figures 1 and 2 was examined using 773 elderly people (mean age 72.6 years (65-97 years)) living in Tokyo, consisting of 332 men and 441 women. The results showed a moderate correlation (correlation coefficient 0.51) with the MMSE, a widely used cognitive function test. The discriminant ability of the device to identify suspected dementia (MMSE<24) was also examined using an ROC curve. The resulting curve shown in Figure 5 showed an area under the curve (AUC): 0.85, sensitivity: 0.81, and specificity: 0.77. This demonstrates that the program and device of the present invention can be used to assess dementia function decline with discriminant ability equivalent to that of the MMSE. [Example 2] (Example of threshold setting) A test was conducted on 455 elderly people living in Tokyo (median age 72 years (65-89 years), 173 men and 282 women) using the dementia function decline likelihood assessment device shown in Figures 1 and 2. Based on the actual data on the occurrence of MCI two years after the test, the predictive ability of the device to predict the occurrence of MCI within two years of the test was examined using an ROC curve. As a result, the curve shown in Figure 6 was obtained, with an area under the curve (AUC): 0.80, sensitivity: 0.81, and specificity: 0.76. From the results obtained, a threshold was calculated using Youden's index. This was used as the threshold for the prediction score.

Claims

1. A computer having a screen and audio output means, a display step of displaying predetermined problems on the screen, the problems being number recitation, number recitation backward, immediate memory, remote memory, a Stroop task, and figure recognition, each of which is randomly generated; an answering step of giving voice instructions to the subject by the voice output means to have the subject answer; a storage step of acquiring the answers given by the subject, determining whether they are correct or incorrect, storing the score, age and gender information on a recording medium, and substituting the score, age and gender information into a predetermined formula to obtain a predicted score, wherein the predetermined formula for obtaining the predicted score is a formula that multiplies each of the number recitation score, number recitation score, immediate memory score, remote memory score, Stroop task score, figure recognition score, gender and age by a predetermined coefficient and adds the result to a predetermined constant; A determination step of comparing the stored predicted score with a predetermined threshold to determine the likelihood of future occurrence of mild cognitive decline; A computer program for causing a computer to execute A computer program characterized in that the predetermined formula in the storage step is the following prediction formula for the occurrence of mild cognitive decline. Prediction formula for mild cognitive decline Predicted score = Number recitation score × (-0.503) + Number recitation score × (-0.540) + Immediate memory score × (-0.056) + Remote memory score × (-0.314) + Stroop task score × (-0.155) + Figure recognition score × (-0.307) + Gender (Male: -0.233, Female: 0) + Age × (0.00036) + Constant (-3.105)

2. The audio output means is a noise-canceling headphone.

2. The computer program according to claim 1.

3. 3. The computer program according to claim 1, wherein in the storing step, when obtaining the predicted score using the predetermined formula, the answer time is also obtained, and the judgment result is corrected by weighting it with the answer time.

4. A dementia function decline likelihood determination device storing the computer program according to claim 1, having a screen and an audio output means, a display means for displaying a predetermined question on the screen; answering means for giving voice instructions to the subject by the voice output means to cause the subject to answer; a storage means for acquiring the answers given by the subject, determining whether the answers are correct or incorrect, and storing the determination result in a recording medium; A dementia function decline likelihood determination device comprising a determination means for substituting the stored result into a predetermined formula, comparing the substitution result with a predetermined threshold, and determining the likelihood of future mild cognitive decline.

Images