Dementia symptom detection device, dementia symptom detection system, dementia symptom detection method, and dementia symptom detection program

Abstract

To provide a dementia symptom detection device capable of detecting a fact that a peripheral symptom is emerging in a dementia patient.SOLUTION: A dementia symptom detection device 20 includes: a storage part that stores a learned model that has learned relationships between a predetermined feature amount on voice information and a plurality of kinds of dementia symptoms; an acquisition part for acquiring voice information on an object person 70; an extraction part for extracting a feature amount from the voice information acquired by the acquisition part; and a determination part for determining whether or not a dementia symptom is emerging in the object person 70 by analyzing the feature amount extracted by the extraction part by the learned model.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a dementia symptom detection device, a dementia symptom detection system, a dementia symptom detection method, and a dementia symptom detection program. [Background technology]

[0002] As the aging of society progresses, the number of dementia patients is also on the rise.

[0003] In this regard, Patent Document 1 below discloses a technology for analyzing a subject's voice to determine signs of dementia. Patent Document 2 below also discloses a technology for analyzing a subject's voice to estimate the presence or absence of mild cognitive impairment. The technologies in Patent Documents 1 and 2 enable early detection of the onset of dementia from the subject's voice. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2021-166663 [Patent Document 2] Japanese Patent Publication No. 2020-127703 Summary of the Invention [Problem to be solved by the invention]

[0005] Patients who have already developed dementia may experience intermittent behavioral and psychological symptoms. In facilities where dementia patients reside, if a dementia patient experiences behavioral and psychological symptoms, the facility staff must quickly recognize that the dementia patient is experiencing these symptoms and take appropriate action.

[0006] The present invention has been made in view of the above-mentioned problems, and therefore, an object of the present invention is to provide a dementia symptom detection device, a dementia symptom detection system, a dementia symptom detection method, and a dementia symptom detection program that are capable of detecting the onset of dementia patient-related symptoms. [Means for solving the problem]

[0007] The above object of the present invention can be achieved by the following means.

[0008] (1) A dementia symptom detection device having a memory unit that stores a trained model that has learned the relationship between predetermined features related to audio information and multiple types of dementia symptoms; an acquisition unit that acquires audio information related to a subject; an extraction unit that extracts the features from the audio information acquired by the acquisition unit; and a judgment unit that determines whether the subject is experiencing dementia symptoms by analyzing the features extracted by the extraction unit using the trained model.

[0009] (2) The dementia symptom detection device described in (1) above, wherein the trained model estimates the type of dementia symptom occurring in the subject.

[0010] (3) The dementia symptom detection device according to (1) or (2) above, wherein the feature is an index relating to the type of voice or an index relating to the degree of voice.

[0011] (4) The dementia symptom detection device according to (1) or (2) above, wherein the feature includes at least one of an index related to volume, an index related to category, an index related to speed, an index related to recognition rate, and an index related to emotion.

[0012] (5) The dementia symptom detection device according to (1) or (2) above, further comprising a notification unit that notifies a terminal device of the determination result by the determination unit.

[0013] (6) A dementia symptom detection device as described in (5) above, wherein, when the trained model estimates that the subject is experiencing two or more types of dementia symptoms, the notification unit notifies the subject of the two or more types of dementia symptoms.

[0014] (7) The dementia symptom detection device according to (1) or (2) above, wherein the dementia symptoms include behavioral and psychological symptoms of dementia.

[0015] (8) A dementia symptom detection system having a detection device and a dementia symptom detection device, wherein the detection device has a microphone that picks up audio related to a subject, and the dementia symptom detection device has: a memory unit that stores a trained model that has learned the relationship between predetermined features related to audio information and multiple types of dementia symptoms; an acquisition unit that acquires audio information output from the microphone; an extraction unit that extracts the features from the audio information acquired by the acquisition unit; and a judgment unit that determines whether the subject is exhibiting dementia symptoms by analyzing the features extracted by the extraction unit using the trained model.

[0016] (9) A dementia symptom detection system as described in (8) above, wherein the detection device further has a camera that photographs the subject, and the microphone is housed in a sensor box together with the camera.

[0017] (10) A method for detecting dementia symptoms, comprising the steps of: (a) preparing a trained model that has learned the relationship between predetermined features related to audio information and multiple types of dementia symptoms; (b) collecting audio related to a subject using a microphone; (c) acquiring audio information output from the microphone; (d) extracting the features from the audio information acquired in (c); and (e) determining whether or not the subject is experiencing dementia symptoms by analyzing the features extracted in (d) using the trained model.

[0018] (11) A dementia symptom detection program executed by a computer having a memory unit, wherein the memory unit stores a trained model that has learned the relationship between predetermined features related to audio information and multiple types of dementia symptoms, and the dementia symptom detection program causes the computer to execute the following steps: (a) acquiring audio information related to a subject; (b) extracting the features from the audio information acquired in (a); and (c) analyzing the features extracted in (b) using the trained model to determine whether the subject is exhibiting the dementia symptoms. [Effects of the Invention]

[0019] According to the present invention, it is possible to detect the occurrence of cognitive impairment symptoms (dementia symptoms) in a dementia patient. [Brief explanation of the drawings]

[0020] Advantages and features provided by one or more embodiments of the present invention will be more fully understood from the following detailed description and the accompanying drawings, which are for purposes of illustration only and are not intended to define the limits of the invention. [Figure 1] FIG. 1 is a diagram illustrating a schematic configuration of a monitoring system. [Figure 2] FIG. 2 is a block diagram showing a schematic configuration of a detection device. [Figure 3] FIG. 2 is a block diagram showing a schematic configuration of a server device. [Figure 4] FIG. 3 is a diagram illustrating the contents stored in a storage unit of the server device. [Figure 5] FIG. 2 is a block diagram showing a schematic configuration of a management terminal device. [Figure 6] FIG. 1 is a block diagram showing a schematic configuration of a mobile terminal device. [Figure 7] 10 is a flowchart showing the procedure of dementia symptom detection processing. [Figure 8] FIG. 10 is a diagram illustrating an example of a voice index. [Figure 9]FIG. 10 is a diagram illustrating an example of an analysis result. [Figure 10] FIG. 1 is a diagram showing an example of behavioral and psychological symptoms. [Figure 11] FIG. 10 is a diagram illustrating an example of a notification screen. DETAILED DESCRIPTION OF THE INVENTION

[0021] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below with reference to the accompanying drawings, but the scope of the present invention is not limited to the disclosed embodiments.

[0022] 1 is a diagram showing the schematic configuration of a monitoring system 1 to which a dementia symptom detection system according to one embodiment of the present invention is applied. In the following, an example will be described in which the dementia symptom detection system according to this embodiment is installed in a building (such as a hospital or elderly care facility) that has multiple rooms, including multiple rooms for multiple subjects and a care station (nurse station).

[0023] 1, the monitoring system 1 includes a plurality of detection devices 10, a server device 20, a management terminal device 30, and a plurality of mobile terminal devices 40. The detection devices 10, the server device 20, the management terminal device 30, and the mobile terminal devices 40 are connected to each other so as to be able to communicate with each other via a communication network 50. The mobile terminal devices 40 are connected to the communication network 50 via an access point 51.

[0024] The detection device 10 is installed, for example, in the room of a subject 70. In the example shown in FIG. 1 , four detection devices 10 are installed in the rooms of subjects 70, namely, A, B, C, and D. The subject 70 is, for example, a dementia patient who has already developed dementia, and the subject 70 may intermittently exhibit behavioral and psychological symptoms of dementia (BPSD), which are peripheral symptoms of dementia. Staff members 80 who provide care, such as nursing care, for the subject 70 each carry a mobile terminal device 40. The management terminal device 30 is used, for example, by an administrator 90 who is a manager who oversees the staff members 80.

[0025] The locations and numbers of the devices included in the monitoring system 1 are not limited to the example shown in Fig. 1. For example, the server device 20 does not have to be located inside the facility building, and may be an external server device connected to the network 50.

[0026] (Detection device 10) 2 is a block diagram showing a schematic configuration of the detection device 10. The detection device 10 is installed as a sensor box on the ceiling or upper part of a wall in the room of the subject 70.

[0027] As shown in FIG. 2, the detection device 10 includes a control unit 11, a communication unit 12, a camera 13, and a microphone 14, which are interconnected by a bus.

[0028] The control unit 11 is configured with a CPU (Central Processing Unit) and memories such as RAM (Random Access Memory) and ROM (Read Only Memory), and controls the above-mentioned units and performs various arithmetic processing according to a program.

[0029] The communication unit 12 is an interface for communicating with other devices such as the server device 20, the management terminal device 30, and the mobile terminal device 40 via the network 50.

[0030] The camera 13 captures an image of the subject 70 from the ceiling or the upper part of the wall of the room, and outputs the captured image (image data).

[0031] The microphone 14 picks up the voice of the subject person 70 in his / her room and outputs voice information (voice data).

[0032] (Server device 20) 3 is a block diagram showing a schematic configuration of the server device 20. The server device 20 corresponds to the dementia symptom detection device of the present invention.

[0033] 3, the server device 20 includes a control unit 21, a communication unit 22, and a storage unit 23, which are interconnected by a bus. Note that, among the above-mentioned units of the server device 20, those having the same functions as the above-mentioned units of the detection device 10 will not be described in order to avoid duplication.

[0034] The storage unit 23 is configured by a hard disk drive (HDD) or a solid state drive (SSD), and stores various programs and various data.

[0035] 4 is a diagram showing the contents stored in the storage unit 23. The storage unit 23 stores a trained model 100 that has learned the relationship between predetermined features (speech indices) related to speech information and multiple types of BPSD. The trained model 100 is a model generated by machine learning using a light gradient boosting machine (LGBM) with nine speech indices (described later) as explanatory variables and the presence or absence of BPSD (present: 1, absent: 0) and the type of BPSD (type numbers 1 to 25) as objective variables.

[0036] The storage unit 23 also stores programs corresponding to the acquisition unit 110, extraction unit 120, determination unit 130, and notification unit 140. The acquisition unit 110 acquires audio information output from the microphone 14. The extraction unit 120 extracts audio indices from the audio information. The determination unit 130 analyzes the audio indices using the trained model 100 to determine whether or not the subject 70 is experiencing BPSD. If the subject 70 is experiencing BPSD, the notification unit 140 notifies the mobile terminal device 40 of this. The functions of each of the above units are fulfilled by the control unit 21 executing the corresponding programs.

[0037] (Management terminal device 30) 5 is a block diagram showing a schematic configuration of the management terminal device 30. The management terminal device 30 is a so-called PC (Personal Computer).

[0038] 5, the management terminal device 30 includes a control unit 31, a communication unit 32, a memory unit 33, a display unit 34, and an input unit 35, which are interconnected by a bus. Note that, among the above-mentioned units of the management terminal device 30, those that have the same functions as the above-mentioned units of the detection device 10 and the server device 20 will not be described in order to avoid duplication.

[0039] The display unit 34 is, for example, a liquid crystal display, and displays various information.

[0040] The input unit 35 includes a keyboard, a numeric keypad, a mouse, etc., and is used to input various types of information.

[0041] (Mobile terminal device 40) 6 is a block diagram showing a schematic configuration of the mobile terminal device 40. The mobile terminal device 40 is a so-called smartphone or tablet terminal.

[0042] 6, the mobile terminal device 40 includes a control unit 41, a storage unit 42, a wireless communication unit 43, and an operation display unit 44, which are interconnected by a bus. Note that, among the above-mentioned units of the mobile terminal device 40, those having the same functions as the above-mentioned units of the detection device 10 and the server device 20 will not be described in order to avoid duplication.

[0043] The wireless communication unit 43 is capable of wireless communication using standards such as Wi-Fi and Bluetooth (registered trademark), and communicates wirelessly with each device via an access point 51 or directly.

[0044] The operation display unit 44 is, for example, a touch panel display, and is used to display and input various types of information.

[0045] According to the monitoring system 1 configured as described above, the voices in the room of the subject 70 are continuously picked up by the microphone 14. The server device 20 then analyzes the voice information output from the microphone 14 and determines whether or not the subject 70 in the room is experiencing BPSD. If it is determined that the subject 70 in the room is experiencing BPSD, this fact is notified to the mobile terminal device 40 of the staff member 80. The operation of the server device 20 will be described below with reference to FIGS. 7 to 11.

[0046] 7 is a flowchart showing the steps of the dementia symptom detection process executed by the server device 20. The process shown in the flowchart in FIG. 7 is executed by the control unit 21 in accordance with a program stored in the storage unit 23 of the server device 20.

[0047] (Step S101) The control unit 21 acquires audio information output from the microphone 14 of the detection device 10. In this embodiment, the control unit 21 acquires one minute's worth of audio information output from the microphone 14. The audio picked up by the microphone 14 includes audio emitted by the subject 70 in the room (such as conversations with the staff 80 or talking to oneself) and audio resulting from the actions of the subject 70 (such as footsteps, the sound of doors opening and closing, and the sound of knocking on the wall).

[0048] (Step S102) The control unit 21 extracts predetermined voice indices from the voice information acquired in the process shown in step S101. In this embodiment, the control unit 21 extracts nine voice indices 200 shown in FIG.

[0049] 8 is a diagram showing an example of the audio index 200. The audio index includes a volume average 201, a volume deviation 202, a category 203, a speed average 204, a speed deviation 205, a recognition rate average 206, a recognition rate deviation 207, an emotional average 208, and an emotional deviation 209.

[0050] The volume average 201 is an index indicating the average value of the volume of the voice for one minute (for example, 45.2 dB), and the volume deviation 202 is an index indicating the standard deviation of the volume of the voice for one minute.

[0051] Category 203 is an index indicating the category into which the transcribed audio content is classified. In this embodiment, the control unit 21 transcribes the conversation, etc. (conversation, monologue) contained in the audio into text, and then classifies the audio content into categories such as (1) dissatisfaction, (2) anger, (3) verbal abuse, and (4) depression based on the words, etc. contained in the text. For example, if the audio content is classified as "anger," a value of "2" is output. The technology for transcribing audio and classifying it is a well-known technology, and therefore a detailed description thereof will be omitted.

[0052] The speed average 204 is an index indicating the average speed of speech etc. contained in one minute of speech (for example, 141.2 / min), and the speed deviation 205 is an index indicating the standard deviation of the speed of speech etc. contained in one minute of speech. The speed of speech etc. can be obtained by transcribing the speech and counting the number of syllables.

[0053] The recognition rate average 206 is an index (for example, 0.81) indicating the average value of the transcription recognition rate (quantification of speech fluency: 0 to 1), and the recognition rate deviation 207 is an index indicating the standard deviation of the recognition rate. When a specific type of BPSD occurs, the recognition rate becomes low. The technology for evaluating the transcription recognition rate is itself a known technology, so a detailed description will be omitted.

[0054] The emotional average 208 is an index indicating the average value of the 20 emotional parameters, and the emotional deviation 209 is an index indicating the standard deviation of the 20 emotional parameters. In this embodiment, the emotional parameters (1) energy, (2) satisfaction, (3) agitation, (4) aggressiveness, (5) stress, (6) uncertainty, (7) excitement, (8) concentration, (9) emotion and cognition, (10) hesitation, (11) brain power, (12) confusion, (13) focused thinking, (14) imaginative activity, (15) extreme emotion, (16) passion, (17) mood, (18) expectation, (19) dissatisfaction, and (20) confidence are quantified, and a value between 0 and 1 (e.g., 0.60) is calculated for each. Then, the average value and standard deviation are calculated for the 20 emotional parameter values. The technology for analyzing a speaker's emotion from speech is itself a well-known technology, so a detailed description will be omitted.

[0055] As described above, in the process shown in step S102, nine voice indices are extracted from the voice information related to the subject 70. Note that the volume average 201, volume deviation 202, speed average 204, speed deviation 205, recognition rate average 206, and recognition rate deviation 207 are indices related to the level of the voice, and the category 203, emotional average 208, and emotional deviation 209 are indices related to the type of voice. Furthermore, if the voice information does not include voices of conversation or monologue, but only everyday sounds such as footsteps, the values ​​of seven of the nine voice indices other than the volume average 201 and volume deviation 202 will be "NULL."

[0056] (Step S103) The control unit 21 analyzes the nine voice indices extracted in the process shown in step S102 using the trained model 100. As described above, the trained model 100 is a model generated by performing machine learning using LGBM with the nine voice indices as explanatory variables and the presence or absence of BPSD (present: 1, absent: 0) and the type of BPSD (type numbers 1 to 25) as objective variables. In this embodiment, the control unit 21 inputs the nine voice indices to the trained model 100, and the trained model 100 estimates the type of BPSD being experienced by the subject 70 based on the nine voice indices.

[0057] Fig. 9 is a diagram showing an example of an analysis result 300 by the trained model 100, and Fig. 10 is a diagram for explaining the types of BPSD shown in Fig. 9. In the analysis result 300, for the 25 types of BPSD shown in Fig. 10, the trained model 100 outputs "1" for symptoms that are estimated to be highly likely to occur in the subject 70, and outputs "0" for symptoms that are estimated to be low in possibility of occurrence.

[0058] As shown in FIG. 9 , in the analysis result 300, for the subject 70 with identification number 001, "1" is output for type number 3 (BPSD_3) and type number 22 (BPSD_22). In other words, the trained model 100 estimates that the subject 70 with identification number 001 is likely to have two types of BPSD: "verbal abuse" and "resistance to care." Furthermore, for the subject 70 with identification number 002, "1" is output for type number 1 (BPSD_1), and the trained model 100 estimates that the subject 70 with identification number 002 is likely to have BPSD of "visual and auditory hallucinations." Similarly, for the subject 70 with identification number 003, "1" is output for type number 2 (BPSD_2), and the trained model 100 estimates that the subject 70 with identification number 003 is likely to have BPSD of "delusions." On the other hand, for subject 70 with identification number 004, "0" is output for all type numbers 1 to 25, and the trained model 100 estimates that subject 70 with identification number 004 is highly likely not experiencing BPSD.

[0059] (Step S104) The control unit 21 determines whether or not BPSD is occurring in the subject 70. If it is determined that BPSD is occurring in the subject 70 (step S104: YES), the control unit 21 proceeds to the processing of step S105. On the other hand, if it is determined that BPSD is not occurring in the subject 70 (step S104: NO), the control unit 21 ends the processing.

[0060] For example, in the analysis result 300 shown in FIG. 9, the control unit 21 determines that the subject 70 with identification number 001 has two types of BPSD: "verbal abuse" and "resistance to care." The control unit 21 also determines that the subject 70 with identification number 002 has a BPSD of "visual and auditory hallucinations." Similarly, the control unit 21 determines that the subject 70 with identification number 003 has a BPSD of "delusion." On the other hand, the control unit 21 determines that the subject 70 with identification number 004 has not developed BPSD.

[0061] (Step S105) The control unit 21 notifies the mobile terminal device 40 that the subject 70 is experiencing BPSD, and ends the process. For example, the control unit 21 notifies the mobile terminal device 40 that the subject 70 with identification number 001 is experiencing two types of BPSD: “abusive language” and “resistance to care.”

[0062] 11 is a diagram showing an example of a notification screen 400. The notification screen 400 is displayed on the operation display unit 44 of the mobile terminal device 40 carried by the staff member 80. For example, if it is determined that the subject 70 is experiencing BPSD of "visual and auditory hallucinations," as shown in FIG. 11, the notification screen 400 is displayed to notify the subject 70 that BPSD is occurring and that the type of BPSD that is occurring is "visual and auditory hallucinations."

[0063] With this configuration, the staff member 80 who sees the notification screen 400 can quickly understand that the subject 70 is experiencing BPSD. In addition, the staff member 80 who sees the notification screen 400 can quickly understand the type of BPSD that the subject 70 is experiencing. Therefore, the staff member 80 can quickly take appropriate action.

[0064] The configuration of the monitoring system 1 described above is a main configuration for explaining the features of the above-mentioned embodiment, but is not limited to the above configuration and can be modified in various ways within the scope of the claims. Furthermore, configurations that are included in general monitoring systems are not excluded.

[0065] For example, in the above embodiment, nine speech indices are used as feature quantities related to speech information. However, the feature quantities are not limited to the above nine speech indices. Only some of the above nine speech indices may be used, or new speech indices may be used.

[0066] In the above embodiment, the case where the subject is experiencing BPSD (brain-related symptoms) as a dementia symptom is described as an example. However, the dementia symptom is not limited to BPSD, and the present invention can be applied to various symptoms related to dementia.

[0067] In the above embodiment, the server device 20 analyzes the voice information of the subject 70 every minute. However, the server device 20 may analyze the voice information of the subject 70 at intervals shorter than one minute. Alternatively, the server device 20 may analyze the voice information of the subject 70 at intervals longer than one minute.

[0068] In the above embodiment, an example has been described in which the server device 20 notifies the mobile terminal device 40 when the subject 70 experiences BPSD. However, the server device 20 may notify the management terminal device 30 or an external display device or the like.

[0069] In the above embodiment, the server device 20 determines whether a subject is experiencing BPSD using a trained model trained by LGBM. However, the machine learning is not limited to LGBM, and trained models trained by multiple regression analysis, deep learning, or the like may be used. Furthermore, the machine learning may be performed for each subject using the subject's past data, or may be performed for each facility using the past data of multiple subjects.

[0070] Furthermore, some or all of the functions of the server device 20 may be provided by the management terminal device 30, the mobile terminal device 40, or the detection device 10.

[0071] Furthermore, the detection device 10, server device 20, management terminal device 30, and mobile terminal device 40 may each be composed of multiple devices, or any of the devices may be included in the other devices to be composed as a single device.

[0072] Furthermore, the means and methods for performing various processes in the monitoring system 1 according to the above-described embodiment can be realized by either a dedicated hardware circuit or a programmed computer. The above program may be provided by a computer-readable recording medium such as a USB (Universal Serial Bus) memory or a DVD (Digital Versatile Disc)-ROM, or may be provided online via a network such as the Internet. In this case, the program recorded on the computer-readable recording medium is typically transferred to and stored in a storage unit such as a hard disk drive. The above program may be provided as standalone application software, or may be incorporated into the software of a device as a function of the device.

[0073] While embodiments of the present invention have been described and illustrated in detail, the disclosed embodiments are made for purposes of illustration and example only and are not intended to be limiting, and the scope of the present invention should be construed by the language of the appended claims. [Explanation of symbols]

[0074] 1. Monitoring system, 10 detection device, 11, 21, 31, 41 control section, 12, 22, 32 Communications Department, 13 cameras, 14 microphones, 20 server equipment, 23,33,42 storage section, 30 management terminal device, 34 Display section, 35 input section, 40 Mobile terminal device, 43 Radio Communications Department, 44 Operation display section, 50 communication networks, 51 access points, 70 Subjects, 80 staff, 90 Administrator, 100 pre-trained models, 200 audio indicators, 300 analysis results, 400 notification screen.

Claims

1. a storage unit that stores a trained model that has learned the relationship between predetermined features related to voice information and multiple types of dementia symptoms; an acquisition unit that acquires voice information about a subject; an extraction unit that extracts the feature amount from the speech information acquired by the acquisition unit; a determination unit that determines whether or not the subject is experiencing symptoms of dementia by analyzing the feature amount extracted by the extraction unit using the trained model; and A dementia symptom detection device having the above.

2. The dementia symptom detection device according to claim 1 , wherein the trained model estimates the type of dementia symptom occurring in the subject.

3. The dementia symptom detection device according to claim 1 , wherein the feature amount is an index relating to a type of voice or an index relating to a degree of voice.

4. The dementia symptom detection device according to claim 1 , wherein the feature amount includes at least one of an index related to volume, an index related to category, an index related to speed, an index related to recognition rate, and an index related to emotion.

5. The dementia symptom detection device according to claim 1 , further comprising a notification unit that notifies a terminal device of a determination result by the determination unit.

6. 6. The dementia symptom detection device according to claim 5, wherein when the trained model estimates that the subject is experiencing two or more types of dementia symptoms, the notification unit notifies the subject of the two or more types of dementia symptoms.

7. The dementia symptom detection device according to claim 1 or 2, wherein the dementia symptom includes behavioral and psychological symptoms of dementia.

8. A dementia symptom detection system having a detection device and a dementia symptom detection device, The detection device includes: A microphone is provided to pick up the voice of the subject, The dementia symptom detection device includes: a storage unit that stores a trained model that has learned the relationship between predetermined features related to voice information and multiple types of dementia symptoms; an acquisition unit that acquires audio information output from the microphone; an extraction unit that extracts the feature amount from the speech information acquired by the acquisition unit; A dementia symptom detection system comprising: a judgment unit that determines whether or not the subject is exhibiting dementia symptoms by analyzing the feature amount extracted by the extraction unit using the trained model.

9. The detection device includes: Further, a camera for photographing the subject is provided. The dementia symptom detection system according to claim 8 , wherein the microphone is housed in a sensor box together with the camera.

10. A step (a) of preparing a trained model that has learned the relationship between predetermined features related to speech information and multiple types of dementia symptoms; (b) collecting a sound related to the subject by a microphone; (c) acquiring audio information output from the microphone; a step (d) of extracting the feature amount from the speech information acquired in the step (c); a step (e) of determining whether or not the subject is experiencing symptoms of dementia by analyzing the feature amount extracted in the step (d) using the trained model; A dementia symptom detection method comprising:

11. A dementia symptom detection program executed by a computer having a storage unit, the storage unit stores a trained model that has trained a relationship between a predetermined feature amount related to voice information and a plurality of types of dementia symptoms; The dementia symptom detection program (a) acquiring audio information about a subject; a step (b) of extracting the feature amount from the speech information acquired in the step (a); a step (c) of determining whether or not the subject is exhibiting a dementia symptom by analyzing the feature amount extracted in the step (b) using the trained model; A dementia symptom detection program that causes the computer to execute the above.

Images