Cognitive function diagnosis support system and program
The cognitive function diagnosis support system accurately evaluates cognitive function by using celebrity images and names, addressing the inefficiencies of conventional methods by leveraging well-known individuals and varying name recognition, resulting in a more objective and user-friendly assessment.
Patent Information
- Application Number
- PCT/JP2024/020486
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-12-12
- Filing Date
- 2024-06-05
- Publication Date
- 2024-12-12
AI Technical Summary
Conventional methods for evaluating cognitive function are time-consuming and inaccurate due to the need for registering photos and names of family members, relatives, and friends, which not all users understand or recognize, and they do not account for variations in user knowledge or the use of well-known individuals.
A cognitive function diagnosis support system and program that displays images of celebrities and determines cognitive function based on correct name recognition, using a storage unit, extraction unit, display unit, input section, determination unit, and result generation unit to evaluate cognitive function through user input, allowing for the use of well-known individuals and varying names such as nicknames or catchphrases, and categorizing by age or activity genre.
Enables more accurate evaluation of cognitive function by leveraging universally recognized celebrities and reducing the effort of registering images, providing a user-friendly and objective assessment of cognitive decline.
Smart Images

Figure JP2024020486_12122024_PF_FP_ABST
Abstract
Description
Cognitive function diagnostic support system and program
[0001] The present invention relates to a system and a program for assisting in the diagnosis of cognitive function.
[0002] Various methods for assessing the level of cognitive function are disclosed, for example, in Patent Literature 1. Patent Literature 1 discloses, as one method for determining the level of cognitive function, an app that registers photos and names of family members, relatives, and friends of users of elderly care facilities, displays one of the photos on the screen, and checks whether the user remembers the name to determine the level of cognitive function of the user.
[0003] JP 2016-71897 A
[0004] However, with the above-mentioned conventional methods, it is necessary to register the photos and names of family members, relatives, and friends for each user in advance, which is time-consuming. Furthermore, it is necessary to obtain the names and photo information as a prerequisite, which is also difficult.
[0005] Therefore, it is preferable to evaluate cognitive function using the name and photo of a person that all users can recognize, and it is desirable to use a well-known person. However, even if a person is well-known, this does not necessarily mean that all users will recognize that person, and the degree of recognition varies from user to user, which creates the problem of not being able to accurately evaluate cognitive function. In addition, it may be desirable to be able to evaluate cognitive function using various images other than people (such as landscapes or famous buildings).
[0006] In consideration of the above problems, the system disclosed below is a cognitive function diagnosis support system that supports the diagnosis of cognitive function, and includes a memory unit that stores data on test images and the names of the test images, an extraction unit that extracts test images from the memory unit, a display unit that displays the test images extracted by the extraction unit, an input unit that accepts answers from subjects, a judgment unit that makes a correct / incorrect judgment based on the degree of agreement between the names included in the answers accepted by the input unit and the names of the test images displayed on the display unit, and a result generation unit that generates an evaluation result regarding cognitive function based on the result of the correct / incorrect judgment.
[0007] Furthermore, the program disclosed below is a cognitive function diagnosis support program that supports the diagnosis of cognitive function, and causes a computer to execute: an extraction process that extracts test images from a storage unit that stores data on test images and names of the person images; a display process that displays the test images extracted in the extraction process; an input process that receives answers from the test subject; a determination process that determines whether the answers are correct or not based on the degree of agreement between the names included in the answers received in the input process and the names of the test images displayed in the display process; and a result generation process that generates an evaluation result regarding cognitive function based on the result of the correctness determination.
[0008] According to the present invention, it is possible to provide a cognitive function diagnostic support system and program that can make a more accurate assessment than conventional systems in assessing cognitive function by matching test images with names.
[0009] 1 is a block diagram showing a schematic configuration of a cognitive function diagnostic support system according to one embodiment; FIG. 2 shows an example of data stored in a question information database 22; FIG. 3 is an example of a screen displayed on the display of a terminal 10 in a cognitive function diagnostic support system 100; FIG. 4 is an example of a question screen; FIG. 5 is an example of a screen displaying an answer input field; FIG. 6 is an example of a screen displaying a determination result; FIG. 7 is an example of a screen showing a history of the correct answer rate; FIG. 8 is a flowchart showing a schematic operation of a cognitive function diagnostic support system according to one embodiment;
[0010] An embodiment of the present disclosure will be described below with reference to the drawings. The present disclosure is not limited to the following embodiment, and appropriate design modifications can be made within the scope of the configuration of the present disclosure. In the following description, the same reference numerals are used in common between different drawings for identical parts or parts having similar functions, and repeated description thereof will be omitted. The configurations described in the embodiment and modified examples may be combined or modified as appropriate. To facilitate understanding of the description, the drawings referred to below show simplified or schematic configurations, and some components may be omitted. Furthermore, in the drawings referred to below, actual images are represented by line drawings.
[0011] [Overview of the Cognitive Function Diagnostic Support System] The cognitive function diagnostic support system according to this embodiment displays images of celebrities and determines the level of cognitive function based on whether the subject can correctly input the name. Note that "cognitive function diagnosis" refers to a diagnosis of dementia and mild cognitive impairment (MCI). The output of this system may be a final evaluation result regarding the cognitive function diagnosis, or may be used as reference information for a doctor or other doctor to make a diagnosis. After displaying an image of a celebrity, this cognitive function diagnostic support system prompts the subject to answer whether or not they know the celebrity. If the subject answers "no," images of other celebrities are displayed. This cognitive function diagnostic support system also uses images of celebrities well known to people of the subject's generation based on their age. Furthermore, this cognitive function diagnostic support system classifies celebrities by activity genre (category) and displays images of celebrities belonging to the category selected by the subject. Furthermore, when a subject inputs an answer, this cognitive function diagnostic support system will recognize it as correct not only when the subject uses the official name (full name) of the celebrity in question, but also when the answer is correct even if it is only the surname, only the first name, a nickname, common name, abbreviation, catchphrase, pen name, or a combination of these, etc. It is desirable that the nickname, common name, abbreviation, catchphrase, etc. be well known by people of that generation.
[0012] In this embodiment, categories are selected by the age group of the subject, but this is not limited to the age group of the subject, and may be by a certain period of time (for example, the 1950s) or by the era name (for example, the Showa era).
[0013] A celebrity refers to a person for whom N1 / (N1+N0) is equal to or greater than a predetermined threshold when a test is conducted to determine whether or not subjects recognize a person image designated in a specific category, with the test subjects being ordinary people (people with normal cognitive functions) of the same age as the test subjects. The total number of test subjects who are determined to recognize the person is N1, and the total number of test subjects who are determined not to recognize the person is N0. This threshold can be arbitrarily selected, and can be set to a value in the range of 0.70 to 0.95, for example, but is not limited to this. The definition of a celebrity shown here is merely an example, and a person who is rated as "known" by a significant number of ordinary people can be considered a "celebrity" here.
[0014] In addition, in the above, in order to define the scope of "famous people," an example was given in which ordinary people of the same age as the test subject were used as test subjects to determine whether they recognized images of people specified in a specific category.However, the test subjects are not limited to ordinary people of the same age as the test subject, and may also be a group consisting of multiple people from an organization, company, organization, etc. to which the test subject belongs (or belonged).
[0015] [Configuration of cognitive function diagnosis support system] Fig. 1 is a block diagram showing a schematic configuration of a cognitive function diagnosis support system 100 according to one embodiment. Note that Fig. 1 shows a schematic functional configuration of the cognitive function diagnosis support system 100, and it is not essential that each of the blocks shown in Fig. 1 exists as hardware. For example, the blocks shown in the control unit 23 are functional blocks realized by the processor of the control unit 23 operating in accordance with a program.
[0016] 1, the cognitive function diagnostic support system 100 includes a terminal 10 used by a subject and a server 20. The terminal 10 accesses the server 20 via a network (not shown) and transmits data input to the terminal 10 to the server 20. The server 20 processes the data transmitted from the terminal 10 and transmits the processing results to the terminal 10. The terminal 10 outputs the processing results received from the server 20. The terminal 10 is realized by a personal computer, a tablet terminal, a smartphone, or the like.
[0017] As shown in FIG. 1 , the terminal 10 includes an input accepting unit 11, a control unit 12, a display 13, and a transceiver unit 14. The input accepting unit 11 includes a keyboard, a touch panel, etc., and accepts input from the subject. The input accepting unit 11 may also accept voice input, in which case it includes a microphone. The control unit 12 includes a processor and controls the overall operation of the terminal 10. The transceiver unit 14 communicates between the terminal 10 and a server 20. The server 20 includes a subject information storage unit 21 that stores information about the subject, a question information database 23 that stores question information, and a control unit 23. The control unit 23, for example, by executing a predetermined program through the processor, realizes a function (extraction unit 23a) of extracting images of celebrities from the question information database 22, a function (determination unit 23b) of determining whether the answer is correct based on the degree of agreement between the name of a person included in the accepted answer and the name of the person image displayed on the display unit, and a function (result generation unit 23c) of generating an assessment result regarding cognitive function based on the result of the correctness determination.
[0018] FIG. 2 shows an example of data stored in the question information database 22. As shown in FIG. 2, the question information database 22 stores images of celebrities to be used in questions, categorized and classified by age. For example, in the category of "Actors and Actresses" and "80s," nine actors and actresses who are well-known to test subjects in their 80s are assigned category numbers 1 through 9, and their images and names are registered. Multiple names considered to be correct answers are also registered for each celebrity. Up to five correct answers, from "Name_1" to "Name_5," can be registered. Correct answers can include a last name (or the first half of a stage name, etc.), a given name (or the second half of a stage name, etc.), a full name, a nickname, a common name, an abbreviation, a catchphrase, etc. A "search query" is a query for extracting an image of the celebrity from the question information database 22.
[0019] The number of celebrities registered per category is not limited to nine. As shown in the example of Figure 2, celebrities may be registered in multiple categories. The number of correct answers registered is not limited to five. When the answer is entered as text, variations in the spelling of the answer (hiragana, katakana, etc.) may be registered. It may also be possible to allow for variations in spelling (for example, "beat" and "beetle"). One or more images of a celebrity may be registered. The image of the celebrity may be a face image or a full-body image.
[0020] Furthermore, images of celebrities may be old images, such as images of the celebrity taken when the subject was a child. In this case, subjects can answer questions about photos they were familiar with in the past, which increases the percentage of subjects who select "I know," making it easier to assess cognitive function. Furthermore, images of celebrities may be not only still images but also videos. In this case, both still images and videos may be used, or only videos may be used.
[0021] [Operation of the Cognitive Function Diagnostic Support System] FIG. 3 shows an example of a screen displayed on the display of the terminal 10 in the cognitive function diagnostic support system 100. As shown in FIG. 3, the display of the terminal 10 displays a screen prompting the subject to select the age group and question category. When the subject selects their own age group and desired question category and clicks the "Question" button, as shown in FIG. 3, the questioning begins. Note that while an example in which the subject selects and inputs their age group is shown here, selecting their age group is optional. For example, when the subject logs in, the age may be automatically determined based on subject information (including date of birth) pre-registered in the subject information storage unit 21 of the cognitive function diagnostic support system 100. Note that when the subject logs in, the subject may be required to input a subject ID and password, or the subject information may be read from a medium readable by the terminal 10. Examples of such a medium include an IC card recording the subject information and a wristband displaying the subject information as a barcode. Alternatively, the subject's biometric information (fingerprint, veins, iris, etc.) may be registered in advance as subject information, eliminating the need to input the information when logging in. Alternatively, face authentication may be used to eliminate the need to input the information when logging in.
[0022] In the example shown in Figure 3, the question categories displayed are "Actors / Actresses," "Singers," "Comedians," and "Politics / Culture / Economy," but the category classification examples are not limited to these, and any type can be set as a category for groups of people including famous people, such as athletes, announcers, novelists, musicians, etc.
[0023] FIG. 4 shows an example of a question screen displayed after "80s" and "Actors & Actresses" are selected and the "Create Question" button is clicked, as shown in FIG. 3 . One or more images of a celebrity are displayed at the top of the screen. When displaying multiple images, the images may be displayed side-by-side, or they may alternate between images at regular intervals or at click timings within the space of a single image. The displayed celebrity image can be enlarged by clicking on the image. The displayed person images are extracted in order from the question information database 22 according to search query Nos. 1 to 9 for "80s" and "Actors & Actresses." Alternatively, the images may be randomly extracted from search query Nos. 1 to 9. Furthermore, while "80s" and "Actors & Actresses" are selected in this embodiment, the selection is not limited to this; multiple categories, such as "80s," "Actors & Actresses," and "Singers," may also be selectable.
[0024] As shown in Figure 4, buttons for "Know" and "Don't Know" are displayed below the person's image. If the subject clicks "Don't Know," the question is skipped and the person image extracted from the search query with the next number is displayed. On the other hand, if the subject clicks "Know," an answer input field (the "Answer Field" shown in Figure 5) is displayed below the person's image, as shown in Figure 5. The subject can enter an answer here. Note that if the entered answer includes a space, the space is excluded from the correctness determination. In other words, if "Yamada Taro" is the correct answer, "Yamada Taro" is also considered correct. When the subject enters a name and presses the "End Answer" button, a new person image extracted from the search query with the next number is displayed. This cycle is repeated until the answer to search query No. 9 is entered. As shown in Figure 5, a field indicating how many questions remain to be answered may also be displayed. In the example of Figure 5, it displays "8 people left."
[0025] On the screen of FIG. 4 , a description of the celebrity may be displayed as a hint along with the person's image. The hint may be displayed after the answer input field is displayed or before the answer input field is displayed. A function for enabling or disabling the display of such a hint when setting the category of FIG. 3 may be provided. A "Hint" button may be provided to manually display the hint rather than automatically. By pressing the "Hint" button, a separate place for displaying the hint may be provided in the blank space of FIG. 4 or FIG. 5 , or the hint may be displayed on a separate screen as a pop-up window. Displaying a hint can shorten the time required to answer or make it easier to recall an answer if the answer has been temporarily forgotten.
[0026] Displaying a hint may affect the assessment result, which will be described later. For example, if the correct answer rate is 60% when a hint is displayed, a downward correction of 5% may be made depending on the number of times the hint is viewed or displayed, or the level of cognitive function may be evaluated in accordance with a two-dimensional table of the correct answer rate and the number of times the hint is viewed or displayed.
[0027] Although not shown, a "reset" button may be provided in addition to "know" and "don't know." It is assumed that even if the subject selects the desired category, most of the images extracted from the search query will not be recognized in the first place. By pressing the "reset" button, the screen shown in Figure 3 will be displayed, allowing the subject to reset the category and start over. The "reset" button may be used to correct an unintentional category selection error, or may be used during the question presentation.
[0028] On the screen of FIG. 4 , if 30 seconds have passed since the image of a person extracted by a certain search query was displayed, it is assumed that "I don't know" was clicked, and a new image of a person extracted by the next search query is displayed. This reduces the time-consuming process of recalling the person's name and reduces the overall system usage time. The time limit is not limited to 30 seconds and may be set arbitrarily. For example, when registering a person's image and name, a response time option (not shown) may be provided to allow the response time to be changed. This eliminates individual differences in the time it takes for the subject to recognize the image and respond. The response time may also be displayed on the screen of FIG. 4 . The display format may be a known format such as an hourglass or digital clock, and is not limited thereto.
[0029] When the subject has completed answering the questions (excluding those who answered "I don't know") to the images of the nine celebrities, the server's control unit 23 compares the input answer with the answers stored in the question information database 22 to determine whether the answer is correct. The server then determines the level of cognitive function according to the percentage of correct answers. The level of cognitive function can be displayed in multiple stages, such as "healthy," "suspected MCI," and "suspected dementia," in descending order of the percentage of correct answers.
[0030] In this embodiment, the process is repeated until an answer to question No. 9 of the search query is entered. However, for example, if the answer selected as "I know" is answered a certain number of times, the process may proceed to determining whether the answer is correct before all questions are completed. Also, after the user is prompted to select "I know" or "I don't know" for the entire search query or a certain number of times, the user may answer all questions for which the user selected "I know."
[0031] Furthermore, if a subject selects "I don't know" a certain number of times, the number of times the subject selected "I don't know" may be taken into account in the assessment result. For example, the certain number may be set to a predetermined number (an integer value), such as 5 times, or may be set to 50% of the entire search query. Alternatively, if the number of times the subject selected "I don't know" is greater than the predetermined number, the accuracy rate may be discounted. For example, even if the accuracy rate is 80%, if the number of times the subject selected "I don't know" exceeds the predetermined number, the accuracy rate may be discounted by, for example, 20% to make it 64%. Furthermore, if a subject has already used the system multiple times and the number of times the subject selected "I don't know" when using the system again deviates from the average number of times the subject previously answered "I don't know" by more than 50%, this information may be taken into account in the assessment result of cognitive function.
[0032] The assessment results may be displayed as a table, as shown in FIG. 6 . In the example shown in FIG. 6 , a total of five questions were asked, and correct answers were given on two of them (the second and fourth questions), resulting in a 40% accuracy rate. The output format of the assessment results is not limited to this and may be arbitrary. For example, while FIG. 6 displays the percentage of correct answers, the level of cognitive function (healthy, suspected MCI, suspected dementia, etc.) determined from the percentage of correct answers may also be displayed. Furthermore, past assessment results for each subject may be stored as a history, and changes over time may be displayed in a graph, as shown in FIG. 7 . The level of cognitive function may also be calculated and displayed based on the change over time (the change in the percentage of correct answers).
[0033] FIG. 8 is a flowchart showing the operation of the cognitive function diagnostic support system 100. This operation is realized by the processor of the control unit 23 of the server 20 executing a program. First, the control unit 23 displays a screen on the display 13 of the terminal 10, as shown in FIG. 3, that prompts the subject to select and input an age and category, and receives the input from the terminal 10 accepted by the input accepting unit of the terminal 10 (step S1). Next, the control unit 23 extracts a portrait image of one celebrity from the question information database 22 according to the age and category selected by the subject, and displays it on the display 13 of the terminal 10 (step S2). Then, when the subject selects the "Know" or "Don't Know" button, if "Know" is selected (Yes in step S3), the process proceeds to step S4, where an answer input field is displayed on the display and an answer input field is accepted from the subject. On the other hand, if "Don't Know" is selected in step S3 (No in step S3), the process returns to step S2.
[0034] After steps S2 to S4 are repeated a predetermined number of times (Yes in step S5), the control unit 23 calculates the correct answer rate based on the predetermined number of answers. The control unit 23 further evaluates the subject's cognitive function level based on the calculated correct answer rate (step S6). The control unit 23 transmits the obtained evaluation result to the terminal 10 and displays it on the display 13 (step S7).
[0035] Although one embodiment for carrying out the present invention has been described above, the following modifications are also possible.
[0036] For example, the time it takes for the subject to input their answer may be measured, and the assessment result may be determined by taking into account not only the accuracy rate but also the input time. The time measured here may be the time from when the image of the celebrity is displayed until the subject finishes inputting, or the time from when the image of the celebrity is displayed until the subject starts inputting. In the former case, the time required for inputting is also taken into consideration. In the latter case, subjects who are capable of cognition but are not good at inputting are less likely to be evaluated as having low cognitive function.
[0037] Furthermore, when the input method is voice input, the time it takes to input a response may be measured, and speech characteristics of the subject may be extracted by determining the pitch and volume of the voice, measuring the interval between utterances, and the like, and these may be taken into consideration when producing a judgment result. Speech characteristics include, for example, insignificant utterances such as "ah" and "hmm." The extraction method of such speech characteristics and their reflection in the judgment result may be publicly known, or a newly constructed trained model using deep learning or reinforcement learning with a neural network may be used. This allows cognitive function to be evaluated not only based on the match between the displayed facial image and name of the celebrity, but also in combination with the speech characteristics, thereby outputting more accurate results.
[0038] In addition, the content of each subject's answers can be recorded as history, and if a subject answers "I don't know" about a person they previously knew, or if they get the name of a person they previously answered correctly incorrectly, this can be reflected in the diagnostic results.
[0039] In the above description, the subject operates the terminal 10 to input the answers, but a caregiver or assistant may listen to the answers from the subject and input them into the terminal 10 on their behalf.
[0040] In the above explanation, a system configured with the terminal 10 and the server 20 was exemplified, but the present invention can also be implemented as an app that runs on a smartphone or tablet terminal. In this case, the processes of steps S1 to S7 described in Figure 8 are each executed by the app. Alternatively, only the question information database may be placed on the network and accessed from the app on the terminal 10.
[0041] In the above example, a screen containing "Know" and "Don't Know" buttons is displayed, and an answer input field is displayed when the "Know" button is clicked, but the "Know" and "Don't Know" buttons and the answer input field may also be displayed on the same screen.
[0042] Furthermore, in the above example, if "I don't know" is selected for one question (image of a person), that question is skipped, and after selecting "I know," the screen transitions to a screen displaying an answer input field as shown in Figure 5. However, it is also possible to have the subject first select "I know" or "I don't know" for all or some of the images corresponding to the selected category, and then present only the images for which "I know" was selected.
[0043] In the above explanation, several examples of parameters related to the diagnosis results have been shown, but these examples may be used alone or in combination. When several parameters are used, weights may be applied to each parameter in consideration of the degree of contribution of each parameter to the cognitive function diagnosis, and artificial intelligence may be used to adjust the weight coefficients.
[0044] Furthermore, in the above description, an example was given in which an image of a famous person was used as the test image. However, the test image is not limited to images of famous people, and any image that is generally considered to be widely known can also be used. Examples of subjects of such images include, for example, landscapes and buildings. More specific examples of landscapes and buildings include, but are not limited to, famous mountains and islands such as Mount Fuji, Sakurajima, and Enoshima, distinctive natural formations such as the Naruto Whirlpools and Niagara Falls, castle ruins such as Himeji Castle and Osaka Castle, scenery of famous places such as the scramble crossing in front of Shibuya Station, ruins such as Machu Picchu and the Colosseum, shrines and temples such as Kiyomizu-dera Temple and Nikko Toshogu Shrine, bronze statues such as the Hachiko statue and the Saigo Takamori statue, Buddhist statues such as the Thousand-Armed Kannon and the Great Buddha of Todai-ji Temple, tower-type buildings such as Tokyo Tower, churches such as the Sagrada Familia and Notre Dame Cathedral, bridges, and skyscrapers. Landscape images of places designated as World Heritage Sites can also be used as test images. Other examples include world-famous paintings such as the "Mona Lisa" and "The Last Supper." Furthermore, "person" can also include fictional characters, such as animation characters (not limited to humans, but may also be animal or robot-like), local characters such as "Yuru-chara" (cute characters), and mascot characters.
[0045] When using images of landscapes, buildings, etc., it is more preferable to use images that can be used to answer names that are proper nouns or that tend to lead to answers that are proper nouns. For example, it is more preferable to use images that have features that evoke names, such as "Tokyo Tower" among tower-type buildings and "Kiyomizu-dera" among shrines and temples. By asking subjects to answer names rather than answering "tower-type buildings," which encompasses multiple buildings, it is possible to more accurately evaluate the consistency between the image and the name answered by the subject.
[0046] Furthermore, the subject may be allowed to select which category of image to use as the test image, from among people, landscapes, buildings, and the like.
[0047] The above disclosure can also be explained as follows: [Configuration 1] A cognitive function diagnostic support system that supports the diagnosis of cognitive function, comprising: a storage unit that stores data on test images and names of the test images, an extraction unit that extracts test images from the storage unit, a display unit that displays the test images extracted by the extraction unit, an input unit that receives answers from a subject, a determination unit that determines whether the answers are correct or incorrect based on the degree of agreement between the names included in the answers received by the input unit and the names of the test images displayed on the display unit, and a result generation unit that generates an evaluation result regarding cognitive function based on the result of the correctness determination.
[0048] According to the above-mentioned configuration 1, a test image extracted from the test images stored in the storage unit is displayed, and the subject is asked to name the subject. The subject's answer is judged as to whether the answer matches the name stored in the storage unit. Furthermore, an evaluation result regarding the subject's cognitive function is generated based on the result of the accuracy judgment. This makes it possible to evaluate cognitive function more accurately than before.
[0049] [Configuration 2] In the configuration 1, the input unit may prompt the subject to answer whether or not they know the names of the test images displayed on the display unit, and may accept input of names only for test images for which the subject answers that they know the names.
[0050] According to this configuration 2, the subject can be asked to answer questions while excluding test images that the subject does not know from the beginning, which makes it possible to more accurately evaluate cognitive function.
[0051] [Configuration 3] In configuration 1 or 2, when the subject inputs an answer to the input unit that he / she does not know the name a number of times exceeding a predetermined criterion, the result generation unit may generate an evaluation result based on the number of times.
[0052] According to the above-mentioned configuration 3, if the number of times the subject answers "I don't know" exceeds the standard even though a test image that the subject is likely to know is displayed, it is possible to show as an evaluation result that cognitive function may be declining.
[0053] [Configuration 4] In configurations 1 to 3, the extraction unit may identify at least one category from among a plurality of categories related to the test image according to a predetermined rule or according to a selection by the subject, and extract test images corresponding to the identified at least one category from the storage unit.
[0054] According to this configuration 4, by having the subject select a category, test images that are likely to be familiar to the subject can be presented, thereby enabling a more accurate assessment of cognitive function.
[0055] [Configuration 5] In any of configurations 1 to 4, it is preferable that the test images include at least one image of a celebrity, a landscape, or a building.
[0056] According to configuration 5, by using test images including images of well-known celebrities to evaluate cognitive function, it is possible to perform the evaluation more objectively. Furthermore, compared to using images of family members, it is possible to save the effort of registering different images for each subject.
[0057] [Configuration 6] In any of Configurations 1 to 5, the determination unit may calculate a rate of correct answers in the correct / incorrect determination, and the evaluation result may further include the rate of correct answers.
[0058] According to the sixth aspect, by presenting the evaluation result including the percentage of correct answers, the evaluation result becomes more objective.
[0059] [Configuration 7] In Configuration 6, the judgment unit may have a timing unit that measures the time from when a test image is displayed on the display unit to when the input unit accepts an answer from the subject, and may evaluate the level of cognitive function based on the time measured by the timing unit and the rate of correct answers.
[0060] According to the seventh aspect, by taking into account the time it takes for the subject to input the answer, it is possible to more accurately evaluate the cognitive function, since generally, as cognitive function declines, the time required to make a judgment tends to increase.
[0061] [Configuration 8] In configurations 1 to 7, the name stored in the storage unit may be the official name, nickname, abbreviation, or common name of the subject of the test image, or a combination thereof.
[0062] According to configuration 8, even if the name input as the answer is not the official name, it can be determined to be the correct answer, so that the cognitive function can be evaluated more reliably.
[0063] [Configuration 9] In configurations 1 to 7, the test image may be an image of a celebrity, and the name stored in the storage unit may be the person's surname, given name, nickname, abbreviation, common name, stage name, catchphrase, pen name, or a combination thereof.
[0064] According to the ninth aspect, even if the name input as the answer is not the official name, it can be regarded as the correct answer, so that the evaluation of cognitive function can be performed more reliably.
[0065] [Configuration 10] In configurations 1 to 9, the system may further include an answer history storage unit that stores historical data of answers given by each subject in association with the subject, and the result generation unit may refer to the historical data stored in the answer history storage unit and generate an evaluation result based on the result of the correct / incorrect judgment and the historical data.
[0066] According to configuration 10, by referring to the historical data of the subject's past answers, it is possible to determine that cognitive function has declined, for example, when the subject is no longer able to answer a name that he or she was previously able to answer correctly.
[0067] [Configuration 11] In configurations 1 to 10, the device may further include a judgment history storage unit that stores historical data on the results of each subject's correct / incorrect judgment in association with the subject, and the result generation unit may refer to the historical data stored in the judgment history storage unit and generate a diagnostic result based on the correct / incorrect judgment results and the historical data.
[0068] According to configuration 11, by evaluating cognitive function based on historical data of the subject's past responses, it becomes possible to determine changes in cognitive function over time.
[0069] [Configuration 12] A cognitive function diagnostic support program that supports the diagnosis of cognitive function, which causes a computer to execute the following steps: an extraction process that extracts test images from a memory unit that stores data on test images and the names of the test images; a display process that displays the test images extracted by the extraction process; an input process that accepts answers from the subject; a determination process that determines whether the answers received by the input process are correct or incorrect based on the degree of agreement between the names included in the answers accepted by the input process and the names of the test images displayed by the display process; and a result generation process that generates evaluation results regarding cognitive function based on the results of the correct or incorrect determination.
[0070] According to the above-mentioned configuration 12, a test image extracted from the test images stored in the storage unit is displayed, and the subject is asked to name the subject. The subject's answer is judged based on the degree of agreement with the name stored in the storage unit. Furthermore, an evaluation result regarding the subject's cognitive function is generated based on the result of the accuracy judgment. This makes it possible to evaluate cognitive function more accurately than before.
[0071] 10: Terminal, 11: Input acceptance unit, 12: Control unit, 13: Display, 14: Transmitting / receiving unit, 20: Server, 21: Subject information storage unit, 22: Question information database, 23: Control unit, 23a: Extraction unit, 23b: Determination unit, 23c: Result generation unit
Claims
1. A cognitive function diagnostic support system that supports the diagnosis of cognitive function, comprising: a memory unit that stores data on test images and names of the test images; an extraction unit that extracts test images from the memory unit; a display unit that displays the test images extracted by the extraction unit; an input unit that accepts answers from subjects; a judgment unit that makes a correct / incorrect judgment based on the degree of agreement between the names included in the answers accepted by the input unit and the names of the test images displayed on the display unit; and a result generation unit that generates evaluation results regarding cognitive function based on the results of the correct / incorrect judgment.
2. The cognitive function diagnostic support system of claim 1, wherein the input unit prompts the subject to answer whether or not he or she knows the names of the test images displayed on the display unit, and accepts input of names only for test images for which the subject answers that he or she knows the names.
3. The cognitive function diagnostic support system according to claim 1, wherein, when the subject inputs into the input unit a number of times that the subject does not know the name, the number of times exceeds a predetermined criterion, the result generation unit generates an evaluation result based on the number of times.
4. The cognitive function diagnostic support system of claim 1, wherein the extraction unit identifies at least one category among a plurality of categories related to the test images according to a predetermined rule or according to a selection by the subject, and extracts test images corresponding to the identified at least one category from the memory unit.
5. The cognitive function diagnostic support system of claim 1, wherein the test images include at least one image of a celebrity, a landscape, or a building.
6. The cognitive function diagnostic support system according to claim 1, wherein the judgment unit calculates a correct answer rate in the correct / incorrect judgment, and the evaluation result further includes the correct answer rate.
7. The cognitive function diagnostic support system of claim 6, wherein the judgment unit has a timing unit that measures the time from when a test image is displayed on the display unit to when the input unit accepts an answer from the subject, and evaluates the level of cognitive function based on the time measured by the timing unit and the rate of correct answers.
8. The cognitive function diagnostic support system of claim 1, wherein the name stored in the memory unit is the official name, nickname, abbreviation, alias, or combination thereof of the subject of the test image.
9. The cognitive function diagnostic support system of claim 1, wherein the test image is an image of a celebrity, and the name stored in the memory unit is the person's surname, first name, nickname, abbreviation, common name, stage name, catchphrase, pen name, or a combination thereof.
10. A cognitive function diagnostic support system as described in claim 1, further comprising an answer history memory unit that stores historical data of answers given by each subject in correspondence with the subject, wherein the result generation unit references the historical data stored in the answer history memory unit and generates an evaluation result based on the result of the correct / incorrect judgment and the historical data.
11. A cognitive function diagnostic support system as described in claims 1 to 10, further comprising a judgment history storage unit that stores historical data on the results of correct / incorrect judgments for each subject in correspondence with the subject, and the result generation unit references the historical data stored in the judgment history storage unit and generates a diagnostic result based on the results of the correct / incorrect judgments and the historical data.
12. A cognitive function diagnostic support program for supporting the diagnosis of cognitive function, the cognitive function diagnostic support program causing a computer to execute the following processes: an extraction process for extracting test images from a memory unit that stores data on test images and the names of the test images; a display process for displaying the test images extracted by the extraction process; an input process for accepting answers from a subject; a determination process for determining whether the answers received by the input process are correct or incorrect based on the degree of agreement between the names included in the answers accepted by the input process and the names of the test images displayed by the display process; and a result generation process for generating evaluation results regarding cognitive function based on the results of the correct or incorrect determination.
Citation Information
Patent Citations
Dementia prevention system
JP2015180933A
Dementia care supporting system, dementia care supporting server and dementia care support program
JP2016071897A
Dementia medical interview device
JP2017049862A
Socket for testing electronics
KR102363837B1
Cognitive training using face-name associations
US20070218441A1
