Assessment system, server device, assessment method, and program

Abstract

An assessment system (2) comprises: a toothbrush (6); an acquisition unit (54) that acquires, from the toothbrush (6), user information relating to a user (4); a user estimation unit (56) that estimates the user (4) on the basis of the user information acquired by the acquisition unit (54); an odor detection unit (18) that is mounted on the toothbrush (6) and detects an odor of the user (4) who is using the toothbrush (6); a processing unit (58) for converting, to a database format, the correspondence relationship between identification information for identifying the user (4) estimated by the user estimation unit (56) and odor information indicating the odor of the user (4) detected by the odor detection unit (18); an assessment unit (62) for assessing, using a trained model for assessing disease risk, the disease risk of the user (4) on the basis of the correspondence relationship converted to the database format by the processing unit (58); and a determination unit (66) for determining that the user (4) who used each of the toothbrush (6) and another device is the same.

Description

Determination system, server device, determination method, and program

[0001] The present disclosure relates to a determination system, a server device, a determination method, and a program.

[0002] 2. Description of the Related Art A medical system for providing optimal treatment to a patient suffering from a disease is known (see, for example, Patent Document 1).

[0003] JP 2022-68362 A

[0004] The conventional medical systems described above only target patients who have actually contracted a disease, and therefore have the problem of being unable to determine the disease risk of users who have not yet contracted the disease.

[0005] The present disclosure provides a determination system, a server device, a determination method, and a program that can easily determine a user's disease risk.

[0006] A determination system according to one aspect of the present disclosure is a determination system for determining a user's disease risk, comprising: a device used by a plurality of first users; an acquisition unit that acquires, from the device, user information relating to each of the first users using the device; a user estimation unit that estimates each of the first users based on the user information acquired by the acquisition unit; an odor detection unit that is mounted on the device and detects an odor of each of the first users using the device; and a correspondence between identification information for identifying each of the first users estimated by the user estimation unit and odor information indicating the odor of each of the first users detected by the odor detection unit. a determination unit that uses a trained model for determining the disease risk to determine the disease risk of each of the first users based on the correspondence relationships databased by the processing unit; and a determination unit that (i) calculates a first index for each of the first users corresponding to the degree of the disease risk determined by the determination unit, (ii) determines a first ranking for a plurality of the first users using the first index, and (iii) determines that a specific first user is the same as a specific second user using the first ranking and a second ranking for a plurality of second users estimated in response to an odor detected by another device.

[0007] Furthermore, a determination system according to one aspect of the present disclosure is a determination system for determining a user's disease risk, comprising: a plurality of devices each used by at least one user; an acquisition unit that acquires user information about the user using each of the devices from each of the devices; a user estimation unit that estimates each of the users based on the user information acquired by the acquisition unit; an odor detection unit that is installed in each of the devices and detects the odor of each of the users using each of the devices; identification information for identifying each of the users estimated by the user estimation unit; and odor information indicating the odor of each of the users detected by the odor detection unit corresponding to each of the devices. a determination unit that uses a trained model for determining the disease risk to determine the disease risk of each of the users for each of the devices based on the first correspondence databased by the processing unit; and a determination unit that (i) calculates, for each of the devices, a first index for each of the users corresponding to the degree of the disease risk determined by the determination unit, (ii) determines, for each of the devices, a user ranking for the multiple users using the first index, and (iii) uses the user ranking determined for each of the devices to determine a second correspondence between each of the users of a specific device among the devices and each of the users of each of the devices other than the specific device.

[0008] These comprehensive or specific aspects may be realized by a system, a method, an integrated circuit, a computer program, or a non-transitory recording medium such as a computer-readable CD-ROM (Compact Disc-Read Only Memory), or may be realized by any combination of a system, a method, an integrated circuit, a computer program, and a non-transitory recording medium.

[0009] According to the determination system etc. according to one aspect of the present disclosure, it is possible to easily determine the disease risk of a user and to determine whether the same user has used different devices.

[0010] 5 is a block diagram showing the configuration of a determination system according to an embodiment. FIG. 6 is a diagram showing an example of a toothbrush used by a user in the determination system according to an embodiment. FIG. 7 is a diagram showing an example of a washing machine used by a user in the determination system according to an embodiment. FIG. 8 is a diagram showing an example of a Western-style toilet device used by a user in the determination system according to an embodiment. FIG. 9 is a flowchart showing the overall operation of the determination system according to an embodiment. FIG. 10 is a flowchart specifically showing the content of step S2 of the flowchart of FIG. 5. FIG. 11 is a flowchart specifically showing the content of step S3 of the flowchart of FIG. 5. FIG. 12 is a flowchart specifically showing the content of step S5 of the flowchart of FIG. 5. FIG. 13 is a flowchart specifically showing the content of step S6 of the flowchart of FIG. 5. FIG. 14 is a flowchart specifically showing the content of step S8 of the flowchart of FIG. 5. FIG. 15 is a flowchart specifically showing the content of step S9 of the flowchart of FIG. 5. FIG. 16 is a flowchart specifically showing the content of step S10 of the flowchart of FIG. 5. FIG. 17 is a flowchart specifically showing the content of step S12 of the flowchart of FIG. 5.

[0011] Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. However, more detailed explanation than necessary may be omitted. For example, detailed explanation of well-known matters or redundant explanation of substantially the same configuration may be omitted. This is to avoid unnecessary redundancy in the following explanation and to facilitate understanding by those skilled in the art.

[0012] The inventors provide the accompanying drawings and the following description to enable those skilled in the art to fully understand the present disclosure, and do not intend for them to limit the subject matter described in the claims.

[0013] (Embodiment) Hereinafter, an embodiment will be described with reference to FIGS.

[0014] [1. Configuration of the Determination System] First, the configuration of a determination system 2 according to an embodiment will be described with reference to Figs. 1 to 4. Fig. 1 is a block diagram showing the configuration of the determination system 2 according to an embodiment. Fig. 2 is a diagram showing an example of a toothbrush 6 used by a user 4 in the determination system 2 according to an embodiment. Fig. 3 is a diagram showing an example of a washing machine 8 used by a user 4 in the determination system 2 according to an embodiment. Fig. 4 is a diagram showing an example of a Western-style toilet device 10 used by a user 4 in the determination system 2 according to an embodiment.

[0015] The determination system 2 is a system for determining the diabetes risk (an example of a disease risk) of each member of a household including the user 4. It is assumed that the household includes one or more members in addition to the user 4.

[0016] As shown in FIG. 1, the determination system 2 includes a toothbrush 6 (an example of an apparatus), a washing machine 8 (an example of an apparatus), a Western-style toilet device 10 (an example of an apparatus), a server device 12, and a learning database 14.

[0017] 2 , the toothbrush 6 is an electric toothbrush installed in a household and used by a user 4. The toothbrush 6 has a camera 16, an odor detection unit 18, and a communication unit 20.

[0018] The camera 16 is an RGB camera disposed at the tip of the toothbrush 6, and captures an image of the inside of the oral cavity of the user 4 while he or she is brushing his or her teeth.

[0019] The odor detection unit 18 is a VOC (Volatile Organic Compounds) sensor located at the tip of the toothbrush 6, and detects odors in the oral cavity of the user 4 while they are brushing their teeth (an example of the odor of the user 4). Specifically, the odor detection unit 18 detects the amount or proportion of volatile sulfur compounds (the odor of periodontal disease bacteria) contained in the odor in the oral cavity of the user 4.

[0020] The communication unit 20 wirelessly communicates with the server device 12. Specifically, the communication unit 20 transmits, to the server device 12, image data showing an image of the inside of the oral cavity of the user 4 captured by the camera 16 and odor information showing the odor in the oral cavity of the user 4 detected by the odor detection unit 18. The odor information is data that quantifies the amount or proportion of volatile sulfur compounds detected by the odor detection unit 18.

[0021] 3, washing machine 8 is a fully automatic washing machine installed in a household and used by multiple members of the household (including user 4). Washing machine 8 has a camera 22, an odor detection unit 24, and a communication unit 26.

[0022] The camera 22 is an RGB camera placed at the loading port 28 of the washing machine 8 and captures images of the clothes 30 of the user 4 that are loaded into the washing machine 8 through the loading port 28.

[0023] Odor detection unit 24 is a VOC sensor disposed inside washing machine 8, and detects the odor of user 4's clothes 30 (an example of user 4's odor) that has been placed into washing machine 8 through loading port 28. Specifically, odor detection unit 24 detects the amount or proportion of ketone bodies (sweet and sour odor) contained in the odor of user 4's clothes 30.

[0024] The communication unit 26 wirelessly communicates with the server device 12. Specifically, the communication unit 26 transmits to the server device 12 image data indicating an image of the clothing 30 of the user 4 captured by the camera 22 and odor information indicating the odor of the clothing 30 of the user 4 detected by the odor detection unit 24. The odor information is data that quantifies the amount or proportion of ketone bodies detected by the odor detection unit 24.

[0025] As shown in Figure 4, the Western-style toilet apparatus 10 is a Western-style toilet installed in a household and is used by multiple members of the household (including a user 4). The Western-style toilet apparatus 10 includes a toilet body 32, a toilet seat 34, a drain box 36, and a drain pipe 38. A water pool 40 is formed inside the toilet body 32. The toilet seat 34 is supported on an opening on the top surface of the toilet body 32 so that it can be opened and closed. The drain box 36 is interposed between the toilet body 32 and the drain pipe 38. When the toilet body 32 is flushed, water collected in the water pool 40 of the toilet body 32 is discharged into the drain pipe 38 via the drain box 36.

[0026] The Western-style toilet device 10 also has an electrocardiograph 42 , odor detection units 44 , 46 , 48 , a urine volume measurement unit 50 , and a communication unit 52 .

[0027] The electrocardiograph 42 is disposed on the toilet seat 34 and measures the electrocardiogram of the user 4 who is seated on (in contact with) the toilet seat 34 .

[0028] The odor detection unit 44 is a VOC sensor located inside the toilet body 32, and detects the odor of the user 4's urine (an example of the user 4's odor) discharged into the toilet puddle 40 of the toilet body 32. Specifically, the odor detection unit 44 detects the amount or proportion of acetaldehyde (the odor produced when sugar alcohol is fermented) contained in the odor of the user 4's urine.

[0029] The odor detection unit 46 is a VOC sensor located inside the drain box 36, and detects the odor of the user 4's urine (an example of the user 4's odor) discharged from the toilet body 32 into the drain box 36. Specifically, the odor detection unit 46 detects the amount or proportion of acetaldehyde contained in the odor of the user 4's urine.

[0030] The odor detection unit 48 is a VOC sensor located inside the drain pipe 38, and detects the odor of the user 4's urine (an example of the user 4's odor) discharged from the drain manhole 36 into the drain pipe 38. Specifically, the odor detection unit 48 detects the amount or proportion of acetaldehyde contained in the odor of the user 4's urine.

[0031] The urination volume measurement unit 50 is a water level sensor disposed in the water pool 40 of the toilet body 32, and measures the amount of urine of the user 4 discharged into the water pool 40 of the toilet body 32 (hereinafter also referred to as "urination volume"). Specifically, the urination volume measurement unit 50 measures the amount of urine of the user 4 based on the amount of change in the water level in the water pool 40 before and after urination.

[0032] The communication unit 52 wirelessly communicates with the server device 12. Specifically, the communication unit 52 transmits to the server device 12 electrocardiogram data indicating the electrocardiogram of the user 4 measured by the electrocardiograph 42, odor information indicating the odor of the urine of the user 4 detected by each of the odor detection units 44, 46, 48, and urination volume information indicating the amount of urine of the user 4 measured by the urination volume measurement unit 50. The odor information is data that quantifies the amount or proportion of acetaldehyde detected by each of the odor detection units 44, 46, 48.

[0033] The server device 12 includes an acquisition unit 54 (an example of a first acquisition unit and a second acquisition unit), a user estimation unit 56, a processing unit 58, a home database 60, a determination unit 62, and an output unit 64.

[0034] The acquisition unit 54 wirelessly communicates with each of the toothbrush 6, the washing machine 8, and the Western-style toilet apparatus 10. Specifically, the acquisition unit 54 acquires (receives) from the communication unit 20 of the toothbrush 6 image data showing an image of the inside of the oral cavity of the user 4 captured by the camera 16 as user information about the user 4 using the toothbrush 6. The acquisition unit 54 also acquires from the communication unit 20 of the toothbrush 6 odor information showing the odor in the oral cavity of the user 4 detected by the odor detection unit 18.

[0035] Acquisition unit 54 also acquires, from communication unit 26 of washing machine 8, image data indicating an image of user 4's clothes 30 taken by camera 22, as user information related to user 4 who has put clothes 30 into washing machine 8. Acquisition unit 54 also acquires, from communication unit 26 of washing machine 8, odor information indicating the odor of user 4's clothes 30 detected by odor detection unit 24.

[0036] The acquisition unit 54 also acquires, from the communication unit 52 of the Western-style toilet device 10, electrocardiogram data indicating the electrocardiogram of the user 4 measured by the electrocardiograph 42, as user information related to the user 4 who is urinating in the Western-style toilet device 10. The acquisition unit 54 also acquires, from the communication unit 52 of the Western-style toilet device 10, odor information indicating the odor of the urine of the user 4 detected by each of the odor detection units 44, 46, 48, and urination volume information indicating the amount of urine excreted by the user 4 measured by the urination volume measurement unit 50.

[0037] When the acquisition unit 54 acquires image data showing an image of the inside of the oral cavity of the user 4 captured by the camera 16, the user estimation unit 56 analyzes feature quantities (e.g., feature quantities indicating periodontal disease) of the image of the inside of the oral cavity of the user 4 included in the image data to estimate the user 4. For example, if the user 4 has periodontal disease, image data showing an image of the inside of the oral cavity of the user 4 with periodontal disease is registered in advance, and the user estimation unit 56 estimates the user 4 based on matching between the pre-registered image data and the image data acquired by the acquisition unit 54. Note that in this specification, "estimating the user 4" means estimating which member of a plurality of members belonging to a household the user 4 corresponds to.

[0038] Furthermore, when the acquisition unit 54 acquires image data showing an image of the clothing 30 of the user 4 captured by the camera 22, the user estimation unit 56 analyzes feature quantities (e.g., feature quantities showing the color, type, etc. of the clothing 30) of the image of the clothing 30 of the user 4 included in the image data, thereby estimating the user 4. For example, by registering image data showing an image of the clothing 30 of the user 4 in advance, the user estimation unit 56 estimates the user 4 based on a match between the pre-registered image data and the image data acquired by the acquisition unit 54.

[0039] Furthermore, when the acquisition unit 54 acquires electrocardiogram data indicating an electrocardiogram measured by the electrocardiograph 42, the user estimation unit 56 analyzes the waveform of the electrocardiogram data to estimate the user 4. For example, by registering electrocardiogram data indicating the electrocardiogram of the user 4 in advance, the user estimation unit 56 estimates the user 4 based on matching between the pre-registered electrocardiogram data and the electrocardiogram data acquired by the acquisition unit 54.

[0040] The processing unit 58 links a personal ID (Identification), which is identification information for identifying the user 4 estimated by the user estimation unit 56, with multiple pieces of odor information detected by the odor detection units 18, 24, 44, 46, and 48, and with urination volume information measured by the urination volume measurement unit 50, thereby creating a database of correspondences (first correspondences) between the personal ID, multiple pieces of odor information, and urination volume information.

[0041] The home database 60 stores correspondence information that is compiled into a database by the processing unit 58 and indicates the correspondence between personal IDs, a plurality of pieces of odor information, and urination volume information.

[0042] The determination unit 62 uses the trained model stored in the training database 14 to determine (infer) the diabetes risk of the user 4 based on the corresponding information stored in the home database 60. Note that if the amount or proportion of volatile organic compounds (e.g., volatile sulfur compounds, ketone bodies, and acetaldehyde) that are secretions specific to diabetic patients and contained in the odor indicated by the odor information is equal to or greater than a predetermined value, and if the amount of urine indicated by the urine volume information is equal to or greater than a predetermined amount, the diabetes risk is determined to be high.

[0043] When the determination unit 62 determines that the user 4 has a high risk of diabetes, the output unit 64 outputs the determination result of the determination unit 62 to, for example, a mobile device or the like of the user 4. At this time, for example, a push notification notifying the user 4 that the user 4 has been determined to have a high risk of diabetes is displayed on the mobile device or the like of the user 4, or an alarm sound is output.

[0044] The training database 14 is located, for example, in a cloud system external to the server device 12. The training database 14 stores a trained model for determining diabetes risk. The trained model is constructed, for example, by performing machine learning using known diabetes risk, odor information, and urination volume information as training data. For example, a neural network, a random forest, a support vector machine, or a self-organizing map is used to construct a logical model in machine learning. Furthermore, the trained model is appropriately updated by machine learning using corresponding information stored in the home database 60 as training input data. Note that, although the training database 14 is located in a cloud system in this embodiment, the present invention is not limited to this and may be located in the server device 12.

[0045] The server device 12 further includes a determination unit 66. The function of the determination unit 66 will be described later.

[0046] [2. Operation of Determination System] [2-1. Overall Operation of Determination System] The overall operation of the determination system 2 according to the embodiment will be described with reference to Fig. 5. Fig. 5 is a flowchart showing the flow of the overall operation of the determination system 2 according to the embodiment.

[0047] 5 , first, user 4 brushes their teeth using toothbrush 6 (S1). Next, user estimation unit 56 of server device 12 estimates user 4 who is brushing their teeth with toothbrush 6 (S2). Next, odor detection unit 18 of toothbrush 6 detects the odor in the oral cavity of user 4 who is brushing their teeth with toothbrush 6 (S3).

[0048] Next, user 4 puts clothes 30 into washing machine 8 (S4). Next, user estimation unit 56 of server device 12 estimates user 4 who puts clothes 30 into washing machine 8 (S5). Next, odor detection unit 24 of washing machine 8 detects the odor of user 4's clothes 30 put into washing machine 8 (S6).

[0049] Next, the user 4 urinates in the Western-style toilet device 10 (S7). Next, the user estimation unit 56 of the server device 12 estimates the user 4 urinating in the Western-style toilet device 10 (S8). Next, the urine volume measurement unit 50 of the Western-style toilet device 10 measures the amount of urine of the user 4 discharged into the Western-style toilet device 10 (S9). Furthermore, each of the odor detection units 44, 46, 48 of the Western-style toilet device 10 detects the odor of the urine of the user 4 discharged into the Western-style toilet device 10 (S10). Note that the order of steps S9 and S10 may be reversed.

[0050] Next, the processing unit 58 of the server device 12 links the personal ID of the user 4 estimated by the user estimation unit 56 with the plurality of pieces of odor information detected by the odor detection units 18, 24, 44, 46, and 48, and the urination volume information measured by the urination volume measurement unit 50 (S11). As a result, the processing unit 58 stores correspondence information, which is a database of correspondence relationships between the personal ID, the plurality of pieces of odor information, and the urination volume information, in the home database 60.

[0051] Finally, the judgment unit 62 of the server device 12 uses the trained model stored in the training database 14 to judge the diabetes risk of the user 4 based on the correspondence information databased by the processing unit 58 (S12).

[0052] In the flowchart of FIG. 5, the user 4 performs step S1 (brushing teeth), step S4 (putting clothes 30 into the washing machine 8), and step S7 (urinating) in that order, but this is not limited to this, and the order of steps S1, S4, and S7 may be arbitrary.

[0053] [2-2. Details of Step S2] Next, details of step S2 in the flowchart of Fig. 5 will be described with reference to Fig. 6. Fig. 6 is a flowchart specifically showing the contents of step S2 in the flowchart of Fig. 5.

[0054] As shown in FIG. 6, first, the camera 16 of the toothbrush 6 captures an image of the inside of the oral cavity of the user 4 who is brushing his / her teeth with the toothbrush 6 (S201).

[0055] Next, the acquisition unit 54 of the server device 12 acquires image data showing an image of the user 4's oral cavity taken by the camera 16 from the communication unit 20 of the toothbrush 6 as user information about the user 4 using the toothbrush 6 (S202).

[0056] Next, the user estimation unit 56 of the server device 12 estimates the user 4 based on the image of the inside of the oral cavity of the user 4 included in the image data acquired by the acquisition unit 54 (S203).

[0057] Next, the user estimation unit 56 determines whether or not a personal ID assigned to the user 4 exists (S204).

[0058] If a personal ID assigned to user 4 exists (YES in S204), the user estimation unit 56 updates the history of user information linked to user 4's personal ID based on the user information acquired in step S202 (S205).

[0059] On the other hand, if there is no personal ID assigned to user 4 (NO in S204), the user estimation unit 56 links and registers the personal ID of user 4 with the user information acquired in step S202 (S206).

[0060] [2-3. Details of Step S3] Next, details of step S3 in the flowchart of Fig. 5 will be described with reference to Fig. 7. Fig. 7 is a flowchart specifically showing the contents of step S3 in the flowchart of Fig. 5.

[0061] As shown in FIG. 7, first, the processing unit 58 of the server device 12 reads out the personal ID of the user 4 who is brushing his / her teeth with the toothbrush 6 from the user estimation unit 56 (S301).

[0062] Next, the odor detection unit 18 of the toothbrush 6 detects the odor in the oral cavity of the user 4 who is brushing his / her teeth with the toothbrush 6 (S302).

[0063] Next, the acquisition unit 54 of the server device 12 acquires odor information indicating the odor in the oral cavity of the user 4 detected by the odor detection unit 18 from the communication unit 20 of the toothbrush 6 (S303).

[0064] Next, the processing unit 58 of the server device 12 estimates whether the current time is before or after tooth brushing and the time zone of the current time based on the date and time information indicating the current date and time (S304). Note that the time used as the criterion for determining whether the current time is before or after tooth brushing may be changed later by the user 4.

[0065] If the current time is before tooth brushing and the current time zone is morning ("before" in S305, "morning" in S306), the processing unit 58 compares the odor information acquired in step S303 with a first threshold (S307). In this case, it is highly likely that user 4 is planning to brush their teeth before breakfast, so the processing unit 58 directly compares the odor information acquired in step S303 with the first threshold. Here, the first threshold is the amount or proportion of volatile sulfur compounds contained in the odor in user 4's oral cavity that can be used to determine whether user 4 is at high risk of diabetes.

[0066] If the odor information is less than the first threshold (YES in S308), the processing unit 58 links the personal ID read in step S301 with the odor information obtained in step S303 and stores them in the home database 60 (S309).

[0067] Returning to step S308, if the odor information is equal to or greater than the first threshold (NO in S308), the processing unit 58 turns on a high oral risk flag (S310). The high oral risk flag indicates that the amount or proportion of volatile sulfur compounds contained in the odor in the oral cavity of the user 4 is equal to or greater than a value that indicates a high risk of diabetes. Next, the processing unit 58 associates the personal ID read in step S301, the odor information acquired in step S303, and the high oral risk flag, and stores them in the home database 60 (S311).

[0068] Returning to step S306, if the current time is before tooth brushing and the current time zone is nighttime ("before" in S305, "night" in S306), it is highly likely that user 4 is about to brush their teeth after dinner, and therefore processing unit 58 detects an odor in user 4's oral cavity other than volatile sulfur compounds (hereinafter referred to as a "background odor") as an initial value (S312). Here, the background odor is, for example, the odor of food or drink consumed by user 4 at dinnertime. Note that the initial value indicating the background odor may be corrected later by user 4. Alternatively, information indicating the mealtime and toothbrushing time may be obtained from an input by user 4 or a log of dietary management app information, and whether the current time is before or after a meal may be estimated based on the obtained information.

[0069] Next, the processing unit 58 compares the odor information obtained by subtracting the background odor from the odor information acquired in step S303 with the first threshold (S313). In this case, since it is highly likely that user 4 is about to brush their teeth after dinner, the processing unit 58 subtracts the background odor from the odor information acquired in step S303 and then compares the odor information with the first threshold. Thereafter, steps S308 to S311 are executed in the same manner as described above.

[0070] Returning to step S305, if the current time is after brushing teeth ("after" in S305), the processing unit 58 calculates the amount of change between the odor information acquired in step S303 and previous odor information after brushing teeth (when the user 4 had a low risk of diabetes), and compares the calculated amount of change with a second threshold (S314). Here, the second threshold is the amount of change from previous data in the amount or proportion of volatile sulfur compounds contained in the odor in the oral cavity of the user 4, which can be used to determine that the risk of diabetes is high. If the user 4 currently has a high risk of diabetes, the odor of volatile sulfur compounds will remain strong in the oral cavity even after brushing teeth, and the calculated amount of change will be equal to or greater than the second threshold.

[0071] If the change in the odor information is less than the second threshold (YES in S315), the processing unit 58 links the personal ID read in step S301 with the odor information obtained in step S303 and stores them in the home database 60 (S309).

[0072] Returning to step S315, if the odor information is equal to or greater than the second threshold (NO in S315), the processing unit 58 turns on the high oral risk flag (S310). Next, the processing unit 58 associates the personal ID read in step S301, the odor information acquired in step S303, and the high oral risk flag, and stores them in the home database 60 (S311).

[0073] [2-4. Details of Step S3] Next, details of step S5 in the flowchart of Fig. 5 will be described with reference to Fig. 8. Fig. 8 is a flowchart specifically showing the content of step S5 in the flowchart of Fig. 5.

[0074] As shown in FIG. 8, first, the camera 22 of the washing machine 8 takes an image of the clothes 30 of the user 4 that have been put into the washing machine 8 (S501).

[0075] Next, the acquisition unit 54 of the server device 12 acquires image data showing an image of the user 4's clothes 30 taken by the camera 22 from the communication unit 26 of the washing machine 8 as user information about the user 4 who put the clothes 30 into the washing machine 8 (S502).

[0076] Next, the user estimation unit 56 of the server device 12 estimates the user 4 based on the image of the clothing 30 of the user 4 included in the image data acquired by the acquisition unit 54 (S503).

[0077] Next, the user estimation unit 56 determines whether or not a personal ID assigned to the user 4 exists (S504).

[0078] If a personal ID assigned to user 4 exists (YES in S504), the user estimation unit 56 updates the history of user information linked to user 4's personal ID based on the user information acquired in step S502 (S505).

[0079] On the other hand, if there is no personal ID assigned to user 4 (NO in S504), the user estimation unit 56 links and registers the personal ID of user 4 with the user information acquired in step S502 (S506).

[0080] [2-5. Details of Step S6] Next, details of step S6 in the flowchart of Fig. 5 will be described with reference to Fig. 9. Fig. 9 is a flowchart specifically showing the content of step S6 in the flowchart of Fig. 5.

[0081] As shown in FIG. 9, first, the processing unit 58 of the server device 12 reads out the personal ID of the user 4 who puts the clothes 30 into the washing machine 8 from the user estimation unit 56 (S601).

[0082] Next, the odor detection unit 24 of the washing machine 8 detects the odor of the clothes 30 of the user 4 that have been put into the washing machine 8 (S602).

[0083] Next, the acquisition unit 54 of the server device 12 acquires odor information indicating the odor of the clothes 30 of the user 4 detected by the odor detection unit 24 from the communication unit 26 of the washing machine 8 (S603).

[0084] Next, the processing unit 58 of the server device 12 compares the odor information acquired in step S603 with a third threshold (S604). Here, the third threshold is the amount or percentage of ketone bodies contained in the odor of the clothing 30 of the user 4, which indicates a high risk of diabetes.

[0085] If the odor information is less than the third threshold (YES in S605), the processing unit 58 links the personal ID read in step S601 with the odor information obtained in step S603 (i.e., the odor information before washing) and stores it in the household database 60 (S606).

[0086] Returning to step S605, if the odor information is equal to or greater than the third threshold (NO in S605), the processing unit 58 turns on the high body odor risk flag (S607). The high body odor risk flag indicates that the amount or proportion of ketone bodies contained in the odor of the clothing 30 of the user 4 is equal to or greater than a value that indicates a high risk of diabetes. Next, the processing unit 58 associates the personal ID read in step S601, the odor information acquired in step S603, and the high body odor risk flag, and stores them in the home database 60 (S608).

[0087] After step S606 or step S608, washing of the clothes 30 begins in the washing machine 8 (S609). Although not shown in the flowchart of FIG. 9 for convenience of explanation, in reality, steps S601 to S608 are repeatedly executed each time each member of the household loads clothes 30 into the washing machine 8 until washing begins in step S609. In this case, the acquired odor information may be, for example, a cumulative value or a difference value from a past value.

[0088] If the laundry was performed using detergent (YES in S610), the processing unit 58 of the server device 12 identifies the type of detergent used based on the odor information indicating the detergent odor from the odor detection unit 24 of the washing machine 8 (S611). After that, the washing of the clothes 30 in the washing machine 8 is completed (S612).

[0089] Next, the odor detection unit 24 of the washing machine 8 detects the odor of the clothes 30 of the user 4 that have been washed in the washing machine 8 (S613).

[0090] Next, the acquisition unit 54 of the server device 12 acquires odor information indicating the odor of the clothes 30 of the user 4 detected by the odor detection unit 24 from the communication unit 26 of the washing machine 8 (S614).

[0091] Next, the processing unit 58 of the server device 12 compares the odor information obtained in step S614 minus the odor information indicating the detergent odor with a fourth threshold (S615). Here, the fourth threshold is the amount or percentage of ketone bodies contained in the odor of the clothing 30 of the user 4, which indicates a high risk of diabetes.

[0092] If the odor information is less than the fourth threshold (YES in S616), the processing unit 58 links the personal ID read in step S601 with the odor information obtained in step S614 minus the odor information indicating the detergent odor (i.e., the odor information after washing using detergent), and stores this in the household database 60 (S617).

[0093] Returning to step S616, if the odor information is equal to or greater than the fourth threshold (NO in S616), the processing unit 58 turns on the high body odor risk flag (S618). Next, the processing unit 58 associates the personal ID read in step S601, the odor information obtained by subtracting the odor information indicating the detergent odor from the odor information acquired in step S614 (i.e., the odor information after washing using the detergent), and the high body odor risk flag, and stores them in the household database 60 (S619).

[0094] Returning to step S610, if washing is performed without using detergent (NO in S610), washing of the clothes 30 in the washing machine 8 is then completed (S620). If user 4 has a high risk of diabetes, there is a high possibility that the smell of ketone bodies is clinging to user 4's clothes 30.

[0095] Next, the odor detection unit 24 of the washing machine 8 detects the odor of the clothes 30 of the user 4 that have been washed in the washing machine 8 (S621).

[0096] Next, the acquisition unit 54 of the server device 12 acquires odor information indicating the odor of the clothes 30 of the user 4 detected by the odor detection unit 24 from the communication unit 26 of the washing machine 8 (S622).

[0097] Next, the processing unit 58 of the server device 12 compares the odor information acquired in step S622 with the fourth threshold value (S623). Then, the process proceeds to step S616.

[0098] If the odor information is less than the fourth threshold (YES in S616), the processing unit 58 links the personal ID read in step S601 with the odor information obtained in step S622 (i.e., the odor information after washing without using detergent) and stores them in the household database 60 (S617).

[0099] Returning to step S616, if the odor information is equal to or greater than the fourth threshold (NO in S616), the processing unit 58 turns on the high body odor risk flag (S618). Next, the processing unit 58 associates the personal ID read in step S601, the odor information acquired in step S622 (i.e., the odor information after washing without detergent), and the high body odor risk flag, and stores them in the household database 60 (S619).

[0100] [2-6. Details of Step S8] Next, details of step S8 in the flowchart of Fig. 5 will be described with reference to Fig. 10. Fig. 10 is a flowchart specifically showing the content of step S8 in the flowchart of Fig. 5.

[0101] As shown in FIG. 10, first, the electrocardiograph 42 of the Western-style toilet device 10 measures the electrocardiogram of the user 4 who is seated on the toilet seat 34 (S801).

[0102] Next, the acquisition unit 54 of the server device 12 acquires electrocardiogram data indicating the electrocardiogram measured by the electrocardiograph 42 from the communication unit 52 of the Western-style toilet device 10 as user information about the user 4 sitting on the toilet seat 34 (S802).

[0103] Next, the user estimation unit 56 of the server device 12 estimates the user 4 based on the electrocardiogram data acquired by the acquisition unit 54 (S803).

[0104] Next, the user estimation unit 56 determines whether or not a personal ID assigned to user 4 exists (S804).

[0105] If a personal ID assigned to user 4 exists (YES in S804), the user estimation unit 56 updates the history of user information linked to user 4's personal ID based on the user information acquired in step S802 (S805).

[0106] On the other hand, if there is no personal ID assigned to user 4 (NO in S804), the user estimation unit 56 links and registers the personal ID of user 4 with the user information acquired in step S802 (S806).

[0107] [2-7. Details of Step S9] Next, details of step S9 in the flowchart of Fig. 5 will be described with reference to Fig. 11. Fig. 11 is a flowchart specifically showing the contents of step S9 in the flowchart of Fig. 5.

[0108] As shown in FIG. 11, first, the processing unit 58 of the server device 12 reads out the personal ID of the user 4 seated on the toilet seat 34 of the Western-style toilet device 10 from the user estimation unit 56 (S901).

[0109] Next, the urination volume measuring unit 50 of the Western-style toilet device 10 measures the urination volume of the user 4 (S902). Next, the acquiring unit 54 of the server device 12 acquires urination volume information indicating the urination volume of the user 4 measured by the urination volume measuring unit 50 from the communication unit 52 of the Western-style toilet device 10 (S903).

[0110] The process waits for 10 seconds after the user 4 sits on the toilet seat 34 (S904), and if the user 4 has not finished urinating (NO in S905), the process returns to step S902 described above.

[0111] On the other hand, if user 4 has completed urination (YES in S905), the processing unit 58 of the server device 12 links the personal ID read in step S901 with the urination volume information acquired in step S903 and stores them in the home database 60 (S906).

[0112] [2-8. Details of Step S10] Next, details of step S10 in the flowchart of Fig. 5 will be described with reference to Fig. 12. Fig. 12 is a flowchart specifically showing the content of step S10 in the flowchart of Fig. 5.

[0113] As shown in FIG. 12, first, the odor detection unit 44 inside the toilet body 32 of the Western-style toilet device 10 detects the odor inside the toilet body 32 before the user 4 urinates as an initial value (S1001).

[0114] Next, the processing unit 58 of the server device 12 reads out the personal ID of the user 4 who is seated on the toilet seat 34 of the Western-style toilet device 10 from the user estimation unit 56 (S1002).

[0115] Next, when the odor detection unit 44 inside the toilet body 32 is used (YES in S1003), the odor detection unit 44 inside the toilet body 32 detects the odor of the urine of the user 4 discharged into the water puddle section 40 of the toilet body 32 (S1004).

[0116] Next, the acquisition unit 54 of the server device 12 acquires odor information indicating the odor of the user 4's urine detected by the odor detection unit 44 from the communication unit 52 of the Western-style toilet device 10 (S1005).

[0117] Next, the processing unit 58 of the server device 12 compares the odor information obtained in step S1005 minus the initial value detected in step S1001 with a fifth threshold (S1006). Here, the fifth threshold is the amount or percentage of acetaldehyde contained in the odor of User 4's urine that indicates a high risk of diabetes.

[0118] If the odor information is less than the fifth threshold (YES in S1007), the processing unit 58 links the personal ID read in step S1002 with the odor information obtained by subtracting the initial value detected in step S1001 from the odor information obtained in step S1005, and stores the linked information in the home database 60 (S1008).

[0119] Returning to step S1007, if the odor information is equal to or greater than the fifth threshold (NO in S1007), the processing unit 58 turns on the high odor risk flag (S1009). The high odor risk flag indicates that the amount or percentage of acetaldehyde contained in the odor of user 4's urine is equal to or greater than a value that indicates a high risk of diabetes. Next, the processing unit 58 associates the personal ID read in step S1002 with the odor information obtained by subtracting the initial value detected in step S1001 from the odor information acquired in step S1005, and the high odor risk flag, and stores them in the home database 60 (S1010).

[0120] Returning to step S1003, if the odor detection unit 46 inside the drain box 36 is used (NO in S1003, YES in S1011), the odor detection unit 46 inside the drain box 36 detects the odor of the urine of user 4 excreted inside the drain box 36 (S1012).

[0121] Next, the acquisition unit 54 of the server device 12 acquires odor information indicating the odor of the user 4's urine detected by the odor detection unit 46 from the communication unit 52 of the Western-style toilet device 10 (S1013).

[0122] Next, the processing unit 58 of the server device 12 compares the odor information acquired in step S1013 with a sixth threshold (S1014). Here, the sixth threshold is the amount or percentage of acetaldehyde contained in the odor of the urine of User 4 that indicates a high risk of diabetes.

[0123] If the odor information is less than the sixth threshold (YES in S1015), proceed to step S1008, and the processing unit 58 links the personal ID read in step S1002 with the odor information obtained in step S1013 and stores them in the home database 60 (S1008).

[0124] Returning to step S1015, if the odor information is equal to or greater than the sixth threshold (NO in S1015), the processing unit 58 turns on the high odor risk flag (S1016). Thereafter, proceeding to step S1010, the processing unit 58 associates the personal ID read in step S1002, the odor information acquired in step S1013, and the high odor risk flag, and stores them in the home database 60 (S1010).

[0125] Returning to step S1003, if the odor detection unit 48 inside the drain pipe 38 is used (NO in S1003, NO in S1011, YES in S1017), the odor detection unit 48 inside the drain pipe 38 detects the odor of the urine of user 4 discharged inside the drain pipe 38 (S1018).

[0126] Next, the acquisition unit 54 of the server device 12 acquires odor information indicating the odor of the user 4's urine detected by the odor detection unit 48 from the communication unit 52 of the Western-style toilet device 10 (S1019).

[0127] Then, the process proceeds to step S1014, where the processing unit 58 of the server device 12 compares the odor information acquired in step S1019 with a sixth threshold value (S1014).

[0128] If the odor information is less than the sixth threshold (YES in S1015), proceed to step S1008, and the processing unit 58 links the personal ID read in step S1002 with the odor information obtained in step S1019 and stores them in the home database 60 (S1008).

[0129] Returning to step S1015, if the odor information is equal to or greater than the sixth threshold (NO in S1015), the processing unit 58 turns on the high odor risk flag (S1016). Thereafter, proceeding to step S1010, the processing unit 58 associates the personal ID read in step S1002, the odor information acquired in step S1019, and the high odor risk flag, and stores them in the home database 60 (S1010).

[0130] Returning to step S1003, if none of the odor detection units 44, 46, 48 will be used (NO in S1003, NO in S1011, NO in S1017), the process of step S10 ends.

[0131] [2-9. Details of Step S11] Next, details of step S11 in the flowchart of FIG. 5 will be described.

[0132] The processing unit 58 of the server device 12 links (a) the personal ID of the user 4 estimated by the user estimation unit 56, (b) the odor information saved in step S309 of Figure 7, (c) the odor information and high oral risk flag saved in step S311 of Figure 7, (d) the odor information saved in steps S606 and S617 of Figure 9, (e) the odor information and high body odor risk flag saved in steps S608 and S619 of Figure 9, (f) the odor information saved in step S1008 of Figure 12, (g) the odor information and high odor risk flag saved in step S1010 of Figure 12, and (h) the urination volume information saved in step S906 of Figure 11.

[0133] Note that the processing unit 58 does not necessarily need to link all of the above (b) to (h) with the above (a), but may link at least one of the above (b) to (h) with the above (a).

[0134] As a result, the processing unit 58 stores correspondence information in the household database 60, which is a database of correspondence relationships between personal IDs, multiple odor information, each flag (high oral risk flag, high body odor risk flag, and high odor risk flag), and urination volume information for each member of the household.

[0135] [2-10. Details of Step S12] Next, details of step S12 in the flowchart of Fig. 5 will be described with reference to Fig. 13 and Fig. 14. Fig. 13 is a flowchart specifically showing the content of step S12 in the flowchart of Fig. 5. Fig. 14 is a diagram for explaining the content of step S12 in the flowchart of Fig. 5.

[0136] As shown in FIG. 13, first, the determination unit 62 of the server device 12 reads out correspondence information for each personal ID of a plurality of members (including user 4) belonging to the household from the household database 60 (S1201).

[0137] When machine learning is performed using the correspondence information ("for learning" in S1202), the determination unit 62 outputs the read correspondence information to the learning database 14 of the cloud system. As a result, the correspondence information read in step S1201 is compiled into a database on the learning database 14 of the cloud system (S1203).

[0138] In the training database 14, machine learning is performed using the correspondence information databased in step S1203 as training input data to create a trained model (S1204). Next, the determination unit 62 reads out the trained model for determining diabetes risk from the training database 14 (S1205).

[0139] If the trained model has not been updated (NO in S1206), the process ends. On the other hand, if the trained model has been updated (YES in S1206), the process proceeds to step S1207, which will be described later.

[0140] Returning to step S1202, if the correspondence information is used to determine (infer) the risk of diabetes ("for inference" in S1202), the determination unit 62 checks the trained model provided from the training database 14 (S1207).

[0141] Next, the determination unit 62 determines the diabetes risk for each individual ID based on the correspondence information using the trained model (S1208). Next, the determination unit 62 records, for each individual ID, the number of times and frequency at which the odor information and urination volume information included in the correspondence information exceeded the respective thresholds corresponding to the diabetes risk (S1209).

[0142] Here, with reference to (a) to (c) of Figure 14, an example of odor information for each of multiple members A, B, C, and D in a case where multiple members A, B, C, and D belong to the same household will be described. In each of the graphs (a) to (c) of Figure 14, the horizontal axis PCA1 represents a person's classification vector, and the vertical axis PCA2 represents a diabetes suspicion vector. PCA2 is one of the indicators that indicates the degree of disease risk as a vector.

[0143] The odor information indicating the odor (X) in the oral cavity of each of multiple members A, B, C, and D is clustered, for example, as shown in (a) of Figure 14. In this case, the odor information corresponding to the personal ID of member A exceeds the threshold corresponding to the risk of diabetes.

[0144] Furthermore, odor information indicating the odor (Y) of each of the clothes of multiple members A, B, C, and D is clustered, for example, as shown in (b) of Figure 14. In this case, multiple pieces of odor information corresponding to the personal ID of member A exceed the threshold corresponding to diabetes risk.

[0145] Furthermore, odor information indicating the urine odor (Z) of each of multiple members A, B, C, and D is clustered, for example, as shown in (c) of Figure 14. In this case, multiple pieces of odor information corresponding to the personal ID of member A exceed the threshold corresponding to diabetes risk.

[0146] Here, members A, B, C, and D for oral odor (X), members A, B, C, and D for clothing odor (Y), and members A, B, C, and D for urine odor (Z) are only clustered into four groups within the same household, so it is impossible to determine whether they are the same person. However, by registering member A, the user of a device (e.g., toothbrush 6) on a smartphone or the like and linking member A's personal ID with oral odor (X), it is possible to estimate through clustering estimation that member A for oral odor (X), member A for clothing odor (Y), and member A for urine odor (Z) are the same person. Similarly, it is possible to estimate that members B, C, and D for oral odor (X), members B, C, and D for clothing odor (Y), and members B, C, and D for urine odor (Z) are the same person. That is, based on the central positions of the four clustered groups, members A, B, C, and D in X, Y, and Z are estimated to be the same person in the order of PCA2 (diabetes suspicion vector).

[0147] In this manner, the server device 12 is provided with a determination unit 66 that determines whether members using multiple devices are the same. The determination unit 66 calculates an index PCA2 (first index) for each of multiple members (multiple first users) estimated in response to an odor detected on a certain device, and determines a first ranking for the multiple members using the index PCA2. Similarly, the determination unit 66 determines a second ranking for multiple members (multiple second users) estimated in response to an odor detected on another device. The determination unit 66 can then use the first ranking and the second ranking to determine whether a specific member (specific first user) estimated on one device is the same as a specific member (specific second user) estimated on another device. For example, if the position of a specific member in the first ranking and the position of a specific member in the second ranking are the same, it can be determined that these members are the same member. Here, "position" means, for example, the order (i.e., the order of arrangement along the vertical axis of each graph in (a) to (c) of FIG. 14).

[0148] More specifically, the determination unit 66 acquires an index PCA2 (second index) corresponding to the degree of disease risk of each of the plurality of members (plurality of second users) obtained by another device, and determines a second ranking using this index PCA2. The determination unit 66 can then determine that a specific second member (specific second user) who exists in the same position in the second ranking as a specific first member (specific first user) in the first ranking is the same as the specific first member.

[0149] In other words, the determination unit 66 (i) calculates, for each device, an index PCA2 (first index) for each member corresponding to the degree of disease risk determined by the determination unit 62, (ii) determines, for each device, a ranking (user ranking) for the multiple members using the index PCA2, and (iii) uses the ranking determined for each device to determine a correspondence relationship (second correspondence relationship) between each member using a specific device among the multiple devices and each member using each device other than the specific device. This correspondence relationship is, for example, a relationship in which a member in a specific order in the ranking of a specific device and a member in the same order as this specific order in the ranking of each of the other devices are the same member.

[0150] In the example shown in (a) to (c) of FIG. 14, among multiple members A, B, C, and D, member A is determined to have a high risk of diabetes.

[0151] 13 , after step S1209, the output unit 64 of the server device 12 outputs the determination result of the determination unit 62 (S1210). For example, if the determination unit 62 determines that member A has a high risk of diabetes, the output unit 64 outputs the determination result of the determination unit 62 to member A's mobile terminal or the like.

[0152] [3. Effects] As described above, in this embodiment, each time user 4 uses toothbrush 6, washing machine 8, and Western-style toilet device 10 in their daily lives, the user 4 is identified and the odor of user 4 is detected. Then, a database is created that contains a correspondence between the personal ID of user 4 and odor information indicating the detected odor of user 4. This makes it possible to collect odor information using the daily actions of user 4, making it easy to determine the diabetes risk of user 4 and to determine whether user 4 who has used different devices is the same user 4.

[0153] Furthermore, if the user 4 is determined to have a high risk of developing diabetes, the user 4 can take appropriate measures to prevent diabetes.

[0154] (Supplementary Note) (Technology 1) A determination system for determining a user's disease risk, comprising: a device used by a plurality of first users; an acquisition unit that acquires user information relating to each of the first users using the device from the device; a user estimation unit that estimates each of the first users based on the user information acquired by the acquisition unit; an odor detection unit that is mounted on the device and detects the odor of each of the first users using the device; and a process of creating a database of correspondence between identification information for identifying each of the first users estimated by the user estimation unit and odor information indicating the odor of each of the first users detected by the odor detection unit. a judgment unit that uses a trained model for determining the disease risk to determine the disease risk of each of the first users based on the correspondence databased by the processing unit; and a judgment unit that (i) calculates a first index for each of the first users corresponding to the degree of the disease risk determined by the judgment unit, (ii) determines a first ranking for a plurality of the first users using the first index, and (iii) determines that a specific first user is the same as a specific second user using the first ranking and a second ranking for a plurality of second users estimated in response to an odor detected by another device.

[0155] According to this system, each time a first user uses a device in daily life, the first user's identity is identified and the first user's odor is detected. A database is then created that stores a correspondence between the first user's identification information and odor information indicating the detected odor of the first user. This allows odor information to be collected using the first user's daily activities, making it easy to determine the first user's disease risk and to determine whether a specific first user and a specific second user who use different devices are the same person.

[0156] (Technology 2) The judgment system described in Technology 1, wherein the judgment unit acquires a second index corresponding to the degree of disease risk of each of the second users obtained by the other device, determines the second ranking using the second index, and judges that the specific second user existing in the second ranking at the same position as the position of the specific first user in the first ranking is the same as the specific first user.

[0157] According to this, by using a first ranking of a specific first user and a second ranking of a specific second user who each use different devices, it can be determined that the specific first user and the specific second user are the same if their positions in the ranking are the same.

[0158] (Technology 3) In the determination system described in Technology 1 or 2, the device is a toothbrush equipped with a camera, the acquisition unit acquires, as the user information, an image of the inside of the oral cavity of each of the first users who is brushing their teeth with the toothbrush, taken by the camera, and the odor detection unit detects an odor in the oral cavity of each of the first users who is brushing their teeth with the toothbrush.

[0159] This allows odor information to be collected each time the first user brushes their teeth with the toothbrush, making it possible to easily determine the first user's risk of disease.

[0160] (Technology 4) The determination system described in Technology 1 or 2, wherein the device is a washing machine having a camera, the acquisition unit acquires, as the user information, images of the clothes of each of the first users taken by the camera and put into the washing machine by each of the first users, and the odor detection unit detects the odor of the clothes of each of the first users put into the washing machine by each of the first users.

[0161] This allows odor information to be collected every time the first user puts clothes into the washing machine, making it easy to determine the first user's risk of illness.

[0162] (Technology 5) The device is a Western-style toilet device having an electrocardiograph, the acquisition unit acquires the electrocardiogram of each of the first users seated in the Western-style toilet device measured by the electrocardiograph as the user information, and the odor detection unit detects the odor of urine of each of the first users discharged into the Western-style toilet device, in a determination system described in Technology 1 or 2.

[0163] This allows odor information to be collected every time the first user urinates in the Western-style toilet device, making it possible to easily determine the first user's risk of disease.

[0164] (Technology 6) The determination system described in Technology 5 further includes a urination volume measuring unit that is mounted on the Western-style toilet device and measures the amount of urine of each of the first users excreted into the Western-style toilet device, and the processing unit creates a database of correspondence between the identification information, the odor information, and urination volume information that indicates the amount of urine of each of the first users measured by the urination volume measuring unit.

[0165] According to this, by taking into consideration the amount of urine of the first user measured by the urine volume measurement unit, the disease risk of the first user can be determined with higher accuracy.

[0166] (Technology 7) The determination system according to any one of Technologies 1 to 6, wherein the disease risk is a diabetes risk.

[0167] This makes it possible to easily determine the first user's risk of diabetes.

[0168] (Technology 8) The determination system according to any one of Technologies 1 to 7, wherein the determination unit uses the trained model to perform a predetermined process on the odor information included in the correspondence databased by the processing unit, thereby determining the disease risk of each of the first users.

[0169] This allows the first user's disease risk to be determined more accurately.

[0170] (Technology 9) A determination system for determining a user's disease risk, comprising: a plurality of devices each used by at least one user; an acquisition unit that acquires user information about the user using each of the devices from each of the devices; a user estimation unit that estimates each of the users based on the user information acquired by the acquisition unit; an odor detection unit that is installed in each of the devices and detects the odor of each of the users using each of the devices; and a database of first correspondences between identification information for identifying each of the users estimated by the user estimation unit and odor information indicating the odor of each of the users detected by the odor detection unit corresponding to each of the devices. a determination unit that uses a trained model for determining the disease risk to determine the disease risk of each of the users for each of the devices based on the first correspondence relationship databased by the processing unit; and a determination unit that (i) calculates, for each of the devices, a first index for each of the users corresponding to the degree of disease risk determined by the determination unit, (ii) determines, for each of the devices, a user ranking for a plurality of the users using the first index, and (iii) uses the user ranking determined for each of the devices to determine a second correspondence relationship between each of the users of a specific device among the devices and each of the users of each of the devices other than the specific device.

[0171] According to this system, each time a user uses a device in daily life, the user's identity is estimated and the user's odor is detected. Then, a first correspondence between the user's identification information and odor information indicating the detected user's odor is stored in a database. This allows odor information to be collected using the user's daily activities, making it possible to easily determine the user's disease risk and determine a second correspondence between users who have used different devices.

[0172] (Technology 10) The determination system described in Technology 9, wherein the determination unit determines, as the second correspondence relationship, that the user in a specific order in the user ranking of the specific device and the user in the same order as the specific order in the user ranking of each of the other devices are the same user.

[0173] According to this, by using the user ranking of users who have used different devices, if the order in the user ranking is the same, it can be determined that the users are the same user.

[0174] (Technology 11) A server device for determining a user's disease risk includes a first acquisition unit that acquires user information about each of a plurality of first users using a device from the device, a user estimation unit that estimates each of the first users based on the user information acquired by the first acquisition unit, a second acquisition unit that acquires odor information that indicates an odor of each of the first users using the device, detected by an odor detection unit installed in the device, and a database that stores correspondence between identification information for identifying each of the first users estimated by the user estimation unit and the odor information acquired by the second acquisition unit. A server device comprising: a processing unit; a determination unit that determines the disease risk of each of the first users based on the correspondence databased by the processing unit using a trained model for determining the disease risk; and a determination unit that (i) calculates a first index for each of the first users corresponding to the degree of disease risk determined by the determination unit, (ii) determines a first ranking for a plurality of the first users using the first index, and (iii) determines that a specific first user is the same as a specific second user using the first ranking and a second ranking for a plurality of second users estimated in response to an odor detected by another device.

[0175] According to this system, each time a first user uses a device in daily life, the first user's identity is identified and the first user's odor is detected. A database is then created that stores a correspondence between the first user's identification information and odor information indicating the detected odor of the first user. This allows odor information to be collected using the first user's daily activities, making it easy to determine the first user's disease risk and to determine whether a specific first user and a specific second user who use different devices are the same person.

[0176] (Technology 12) A server device for determining a user's disease risk includes a first acquisition unit that acquires, from each of a plurality of devices, user information about the users using each of the devices, each of which is used by at least one user; a user estimation unit that estimates each of the users based on the user information acquired by the first acquisition unit; a second acquisition unit that acquires odor information that indicates an odor of each of the users using each of the devices, the odor information being detected by an odor detection unit installed in each of the devices; and a database that stores a first correspondence between identification information for identifying each of the users estimated by the user estimation unit and the odor information acquired by the second acquisition unit. a determination unit that uses a trained model for determining the disease risk to determine the disease risk of each of the users for each of the devices based on the first correspondence relationship databased by the processing unit; and a determination unit that (i) calculates, for each of the devices, a first index for each of the users corresponding to the degree of the disease risk determined by the determination unit, (ii) determines, for each of the devices, a user ranking for a plurality of the users using the first index, and (iii) uses the user ranking determined for each of the devices to determine a second correspondence relationship between each of the users of a specific device among the devices and each of the users of each of the devices other than the specific device.

[0177] According to this system, each time a user uses a device in daily life, the user's identity is estimated and the user's odor is detected. Then, a first correspondence between the user's identification information and odor information indicating the detected user's odor is stored in a database. This allows odor information to be collected using the user's daily activities, making it possible to easily determine the user's disease risk and determine a second correspondence between users who have used different devices.

[0178] (Technology 13) A method for determining a user's risk of disease, the method comprising: (a) acquiring user information about a plurality of first users using a device from the device; (b) estimating each of the first users based on the user information acquired in (a); (c) detecting the odor of each of the first users using the device using an odor detection unit mounted on the device; and (d) creating a database of correspondence between identification information for identifying each of the first users estimated in (b) and odor information indicating the odor of each of the first users detected in (c). (e) using a trained model for determining the disease risk, determining the disease risk of each of the first users based on the correspondence databased in (d); and (f) calculating a first index for each of the first users corresponding to the degree of disease risk determined in (e), determining a first ranking for a plurality of the first users using the first index, and determining that a specific first user is the same as a specific second user using the first ranking and a second ranking for a plurality of second users estimated in response to an odor detected by another device.

[0179] According to this system, each time a first user uses a device in daily life, the first user's identity is identified and the first user's odor is detected. A database is then created that stores a correspondence between the first user's identification information and odor information indicating the detected odor of the first user. This allows odor information to be collected using the first user's daily activities, making it easy to determine the first user's disease risk and to determine whether a specific first user and a specific second user who use different devices are the same person.

[0180] (Technology 14) A method for determining a user's risk of disease, comprising: (a) acquiring, from each of a plurality of devices, user information relating to the users using each of the devices, each of which is used by at least one user; (b) estimating each of the users based on the user information acquired in (a); (c) detecting an odor of each of the users using each of the devices using an odor detection unit mounted on each of the devices; and (d) creating a database of first correspondences between identification information for identifying each of the users estimated in (b) and odor information indicating the odor of each of the users detected in (c) corresponding to each of the devices. (e) using a trained model for determining the disease risk, determine the disease risk of each of the users for each of the devices based on the first correspondence relationship databased in (d); (f) for each of the devices, obtain a first index for each of the users corresponding to the degree of disease risk determined in (e); (g) for each of the devices, determine a user ranking for a plurality of the users using the first index; and (h) using the user ranking determined for each of the devices, determine a second correspondence relationship between each of the users of a specific device among the devices and each of the users of each of the other devices other than the specific device.

[0181] According to this system, each time a user uses a device in daily life, the user's identity is estimated and the user's odor is detected. Then, a first correspondence between the user's identification information and odor information indicating the detected user's odor is stored in a database. This allows odor information to be collected using the user's daily activities, making it possible to easily determine the user's disease risk and determine a second correspondence between users who have used different devices.

[0182] (Technology 15) A program that causes a computer to execute the determination method according to Technology 13 or 14.

[0183] (Other Modifications, etc.) As described above, the above-described embodiments have been described as examples of the technology disclosed in the present application. However, the technology in the present disclosure is not limited to these, and can be applied to embodiments in which modifications, substitutions, additions, omissions, etc. are made as appropriate. Furthermore, it is also possible to combine the components described in the above-described embodiments to create new embodiments.

[0184] Therefore, other embodiments will be exemplified below.

[0185] In the above embodiment, the determining unit 62 determines the risk of diabetes as the disease risk, but is not limited to this, and may determine, for example, the risk of hypertension, the risk of obesity, or the risk of dyslipidemia.

[0186] In the above embodiment, the determination unit 62 uses a trained model to determine the disease risk of the user 4 based on the odor information included in the correspondence information, but this is not limited to this. The determination unit 62 may use a trained model to perform a predetermined process (e.g., calculate a difference value of change over time) on the odor information included in the correspondence information to determine the disease risk of the user 4.

[0187] In the above embodiment, the devices are the toothbrush 6, the washing machine 8, and the Western-style toilet device 10, but the devices are not limited to these and may be any other home appliances.

[0188] In the above-described embodiments, each component may be configured with dedicated hardware, or may be realized by executing a software program suitable for each component. Each component may be realized by a program execution unit such as a CPU or processor reading and executing a software program recorded on a non-transitory recording medium such as a hard disk or semiconductor memory.

[0189] Furthermore, some or all of the functions of the determination system 2 according to each of the above-described embodiments may be realized by a processor such as a CPU executing a program.

[0190] As described above, the embodiments have been described as examples of the technology in the present disclosure, and for that purpose, the accompanying drawings and detailed description have been provided.

[0191] Therefore, the components shown in the accompanying drawings and detailed description may include not only essential components for solving the problem, but also components that are not essential for solving the problem in order to illustrate the above technology. Therefore, the fact that these non-essential components are shown in the accompanying drawings or detailed description should not be interpreted as immediately indicating that these non-essential components are essential.

[0192] Furthermore, since the above-described embodiments are intended to illustrate the technology of the present disclosure, various modifications, substitutions, additions, omissions, etc. may be made within the scope of the claims or their equivalents.

[0193] The present disclosure is applicable as a determination system for determining a user's disease risk, etc.

[0194] 2 Determination system 4 User 6 Toothbrush 8 Washing machine 10 Western-style toilet device 12 Server device 14 Learning database 16, 22 Camera 18, 24, 44, 46, 48 Odor detection unit 20, 26, 52 Communication unit 28 Insertion slot 30 Clothes 32 Toilet body 34 Toilet seat 36 Drainage manhole 38 Drainage pipe 40 Water puddle 42 Electrocardiograph 50 Urine volume measurement unit 54 Acquisition unit 56 User estimation unit 58 Processing unit 60 Home database 62 Determination unit 64 Output unit 66 Judgment unit

Claims

1. A determination system for determining a user's disease risk, comprising: a device used by a plurality of first users; an acquisition unit that acquires user information relating to each of the first users using the device from the device; a user estimation unit that estimates each of the first users based on the user information acquired by the acquisition unit; an odor detection unit that is mounted on the device and detects the odor of each of the first users using the device; a processing unit that creates a database of correspondence between identification information for identifying each of the first users estimated by the user estimation unit and odor information indicating the odor of each of the first users detected by the odor detection unit; and a determination unit that uses a trained model for determining the disease risk to determine the disease risk of each of the first users based on the correspondence databased by the processing unit. A determination system comprising: (i) a determination unit that calculates a first index for each of the first users corresponding to the degree of disease risk determined by the determination unit; (ii) determines a first ranking for a plurality of the first users using the first index; and (iii) determines that a specific first user is the same as a specific second user using the first ranking and a second ranking for a plurality of second users estimated in response to an odor detected by another device.

2. The judgment system described in claim 1, wherein the judgment unit obtains a second index corresponding to the degree of disease risk of each of the second users obtained by the other device, determines the second ranking using the second index, and judges that the specific second user existing in the same position in the second ranking as the position of the specific first user in the first ranking is the same as the specific first user.

3. The determination system described in claim 1, wherein the device is a toothbrush having a camera, the acquisition unit acquires an image of the oral cavity of each of the first users photographed by the camera while they are brushing their teeth with the toothbrush as the user information, and the odor detection unit detects an odor in the oral cavity of each of the first users while they are brushing their teeth with the toothbrush.

4. The determination system described in claim 1, wherein the device is a washing machine having a camera, the acquisition unit acquires images of the clothes of each of the first users that were taken by the camera and placed into the washing machine by each of the first users as the user information, and the odor detection unit detects the odor of the clothes of each of the first users that were placed into the washing machine by each of the first users.

5. The determination system described in claim 1, wherein the device is a Western-style toilet device having an electrocardiogram, the acquisition unit acquires the electrocardiogram of each of the first users seated in the Western-style toilet device measured by the electrocardiogram as the user information, and the odor detection unit detects the odor of urine of each of the first users discharged into the Western-style toilet device.

6. The determination system according to claim 5, further comprising a urination volume measuring unit mounted on the Western-style toilet device and measuring the amount of urine of each of the first users discharged into the Western-style toilet device, and the processing unit creates a database of correspondence between the identification information, the odor information, and urination volume information indicating the amount of urine of each of the first users measured by the urination volume measuring unit.

7. The judgment system according to any one of claims 1 to 6, wherein the disease risk is a diabetes risk.

8. A determination system as claimed in any one of claims 1 to 6, wherein the determination unit uses the learned model to perform a predetermined processing on the odor information contained in the correspondence databased by the processing unit, thereby determining the disease risk of each of the first users.

9. A determination system for determining a user's disease risk, comprising: a plurality of devices, each used by at least one user; an acquisition unit that acquires user information relating to the user using each of the devices from each of the devices; a user estimation unit that estimates each of the users based on the user information acquired by the acquisition unit; an odor detection unit that is mounted on each of the devices and detects the odor of each of the users using each of the devices; a processing unit that creates a database of a first correspondence between identification information for identifying each of the users estimated by the user estimation unit and odor information indicating the odor of each of the users detected by the odor detection unit corresponding to each of the devices; and a determination unit that uses a trained model for determining the disease risk to determine the disease risk of each of the users for each of the devices, based on the first correspondence databased by the processing unit. (i) for each of the devices, determining a first index for each of the users corresponding to the degree of disease risk determined by the determination unit; (ii) for each of the devices, determining a user ranking for the multiple users using the first index; and (iii) using the user ranking determined for each of the devices, determining a second correspondence relationship between each of the users of a specific one of the devices and each of the users of each of the other devices other than the specific device.

10. The determination system described in claim 9, wherein the determination unit determines, as the second correspondence relationship, that the user in a specific order in the user ranking of the specific device and the user in the same order as the specific order in the user ranking of each of the other devices are the same user.

11. A server device for determining a user's disease risk, comprising: a first acquisition unit that acquires user information on each of a plurality of first users using a device from the device; a user estimation unit that estimates each of the first users based on the user information acquired by the first acquisition unit; a second acquisition unit that acquires odor information indicating an odor of each of the first users using the device, detected by an odor detection unit mounted on the device; a processing unit that creates a database of correspondence between identification information for identifying each of the first users estimated by the user estimation unit and the odor information acquired by the second acquisition unit; and a determination unit that uses a trained model for determining the disease risk to determine the disease risk of each of the first users based on the correspondence databased by the processing unit. A server device comprising: (i) a judgment unit that calculates a first indicator for each of the first users corresponding to the degree of disease risk judged by the judgment unit; (ii) determines a first ranking for the multiple first users using the first indicator; and (iii) determines that a specific first user is the same as a specific second user using the first ranking and a second ranking for multiple second users estimated in response to an odor detected by another device.

12. A server device for determining a user's disease risk, comprising: a first acquisition unit that acquires, from each of a plurality of devices, each of which is used by at least one user, user information relating to the user using the respective device; a user estimation unit that estimates each of the users based on the user information acquired by the first acquisition unit; a second acquisition unit that acquires odor information indicating an odor of each of the users using each of the devices, detected by an odor detection unit mounted on each of the devices; a processing unit that creates a database of a first correspondence between identification information for identifying each of the users estimated by the user estimation unit and the odor information acquired by the second acquisition unit; and a determination unit that uses a trained model for determining the disease risk to determine the disease risk of each of the users for each of the devices, based on the first correspondence databased by the processing unit. (i) for each of the devices, determining a first index for each of the users corresponding to the degree of disease risk determined by the determination unit; (ii) for each of the devices, determining a user ranking for the multiple users using the first index; and (iii) using the user ranking determined for each of the devices, determining a second correspondence relationship between each of the users of a specific one of the devices and each of the users of each of the other devices other than the specific device.

13. A method for determining a user's disease risk, comprising: (a) acquiring user information on a first user from a plurality of first users who are using a device; (b) estimating each of the first users based on the user information acquired in (a); (c) detecting an odor of each of the first users using the device using an odor detection unit mounted on the device; (d) creating a database of correspondence between identification information for identifying each of the first users estimated in (b) and odor information indicating the odor of each of the first users detected in (c); (e) using a trained model for determining the disease risk, determining the disease risk of each of the first users based on the correspondence databased in (d); (f) calculating a first indicator for each of the first users corresponding to the degree of disease risk determined in (e), determining a first ranking for the multiple first users using the first indicator, and determining that a specific first user is the same as a specific second user using the first ranking and a second ranking for multiple second users estimated in response to an odor detected by another device.

14. A method for determining a user's disease risk, comprising: (a) acquiring, from each of a plurality of devices, user information relating to the user using each of the devices, each of which is used by at least one user; (b) estimating each of the users based on the user information acquired in (a); (c) detecting the odor of each of the users using each of the devices using an odor detection unit mounted on each of the devices; (d) creating a database of a first correspondence between identification information for identifying each of the users estimated in (b) and odor information indicating the odor of each of the users detected in (c) corresponding to each of the devices; (e) using a trained model for determining the disease risk, determining the disease risk of each of the users for each of the devices based on the first correspondence databased in (d); (f) calculating, for each of the devices, a first index for each of the users corresponding to the degree of the disease risk determined in (e); (g) determining, for each of the devices, a user ranking for the multiple users using the first index; and (h) using the user ranking determined for each of the devices, determining a second correspondence relationship between each of the users of a specific device among the devices and each of the users of other devices other than the specific device.

15. A program for causing a computer to execute the determination method according to claim 13 or 14.

Images