Health Information Prediction System and Information Prediction System

Abstract

The health information estimation system according to the implementation method includes: a data acquisition unit that acquires data detected by sensors configured in the water space; a data construction determination unit that determines a data construction based on a data arrangement formed by associating the data acquired by the data acquisition unit with date and time information; and a health information estimation unit that estimates the user's health information based on the data arrangement, wherein the health information estimation unit selects a model corresponding to the data construction from a plurality of health information estimation models to estimate the user's health status.

Description

Technical Field

[0001] The embodiments of the present invention relate to a health information estimation system and an information estimation system. Background Technology

[0002] Conventionally, a technology has been provided for collecting data in water-using spaces such as toilets through sensor detection. For example, a toilet seat device has been provided that includes a gas sensor that can detect excrement gases (such as flatulence) emitted when a toilet user (hereinafter also referred to as "user") defecates (see, for example, Patent Documents 1 and 2). In addition, a system has been provided that determines the user's health information based on excrement data obtained by sensors installed in the toilet (see, for example, Patent Document 3).

[0003] Patent Document 1: Japanese Patent Application Publication No. 2005-315836

[0004] Patent Document 2: Japanese Patent Application Publication No. 2016-145806

[0005] Patent Document 3: US Patent No. 11881313 Summary of the Invention

[0006] However, the aforementioned prior art has room for improvement. For example, while the prior art can use data obtained from water-using spaces such as toilets to estimate the health of users, it is sometimes difficult to appropriately estimate health-related information when the available data fluctuates or when the data used in the processing changes. Furthermore, there is room for improvement regarding how to estimate health-related information when there is a wide variety of available data. Therefore, it is desirable, for example, to appropriately estimate health-related information in accordance with the data used in the processing.

[0007] In light of the above, the appropriate inference of information related to human health has become a topic of discussion.

[0008] The purpose of the embodiments of the present invention is to provide a health information estimation system and an information estimation system capable of appropriately estimating information related to human health.

[0009] One embodiment of the health information estimation system is characterized by comprising: a data acquisition unit that acquires data detected by sensors configured in the water space; a data construction determination unit that determines a data construction based on a data arrangement formed by associating the data acquired by the data acquisition unit with date and time information; and a health information estimation unit that estimates the user's health information based on the data arrangement, wherein the health information estimation unit selects a model corresponding to the data construction from a plurality of health information estimation models to estimate the user's health status.

[0010] According to one embodiment, a health information estimation system determines a data structure based on a data arrangement related to data detected by sensors deployed in the water usage space. Correspondingly, a model corresponding to the data structure is selected from multiple health information estimation models to estimate the user's health status, thereby enabling the estimation of the user's health information based on the acquired data. For example, based on the structure of data obtained from sensors deployed in the water usage space, the health information estimation system selects the optimal health information estimation model from multiple health information estimation models, thereby providing personalized health information tailored to the user's condition. Thus, the health information estimation system can appropriately estimate information related to a person's health.

[0011] One embodiment of the health information estimation system is characterized by comprising: a first data acquisition unit that acquires data detected by sensors configured in the water space; a second data acquisition unit that acquires external data; a data construction determination unit that determines a data construction based on a data arrangement formed by associating at least one of the data acquired by the first data acquisition unit and the second data acquisition unit according to date and time information; and a health information estimation unit that estimates the user's health information based on the data arrangement, wherein the health information estimation unit selects a model corresponding to the data construction from a plurality of health information estimation models to estimate the user's health status.

[0012] According to one embodiment, a health information estimation system estimates a user's health status based on a data structure determined by an arrangement of data related to data detected by sensors deployed in the water-using space or data obtained from external sources. Correspondingly, the system selects a model from multiple health information estimation models that corresponds to the data structure to estimate the user's health status, thereby enabling the estimation of the user's health information based on the acquired data. For example, based on the structure of data obtained from sensors deployed in the water-using space, the system selects the optimal health information estimation model from multiple health information estimation models, thereby providing personalized health information tailored to the user's condition. Thus, the health information estimation system can appropriately estimate information related to a person's health.

[0013] One embodiment of the health information estimation system is characterized by comprising: a first data acquisition unit that acquires data detected by sensors configured in the toilet space; a second data acquisition unit that acquires external data; a data construction determination unit that determines a data construction based on a data arrangement formed by associating at least one of the data acquired by the first data acquisition unit and the second data acquisition unit according to date and time information; and a health information estimation unit that estimates the user's health information based on the data arrangement, wherein the health information estimation unit selects a model corresponding to the data construction from a plurality of health information estimation models to estimate the user's health status.

[0014] According to one embodiment, a health information estimation system estimates the user's health status based on a data structure determined by an arrangement of data related to data detected by sensors installed in the water-using space (i.e., the toilet space) or data obtained from external sources. Correspondingly, the system selects a model from multiple health information estimation models that corresponds to the data structure to estimate the user's health status, thereby estimating the user's health information in accordance with the acquired data. Thus, the health information estimation system can appropriately estimate information related to a person's health. For example, based on the structure of data obtained from sensors installed in the water-using space, the system selects the optimal health information estimation model from multiple health information estimation models, thereby providing personalized health information tailored to the user's condition.

[0015] In one embodiment of the health information estimation system, the data construction decision unit determines the data construction based on the data arrangement formed by associating the data obtained by the data acquisition unit with personal information or date information.

[0016] According to one embodiment, the health information estimation system determines the data structure based on the arrangement of data associated with personal information or date information, thereby enabling appropriate determination of the data structure. Thus, the health information estimation system can appropriately estimate information related to a person's health. For example, by associating the acquired data with date information or personal information, the health information estimation system can appropriately provide health information even when multiple users are present.

[0017] In a health information estimation system according to one embodiment, the data construction determination unit determines the data construction based on a combination of at least one data in the data arrangement.

[0018] According to one embodiment, the health information estimation system determines the data structure based on the combination of data arranged in a data arrangement, thereby enabling the appropriate determination of the data structure. Thus, the health information estimation system can appropriately estimate information related to a person's health. For example, even if the data arrangement is incomplete, the health information estimation system can still provide health information corresponding to all combinations of data.

[0019] In one embodiment of the health information estimation system, the data construction decision unit supplements the missing data arrangement in the latest data arrangement with past data arrangements within a specified date and time range, and determines the data construction based on the supplemented data arrangement.

[0020] According to one embodiment, the health information estimation system supplements missing data arrangement and determines the data structure based on the supplemented data arrangement, thereby enabling the appropriate determination of the data structure. Thus, the health information estimation system can appropriately estimate information related to a person's health. For example, even if there are missing data arrangement, the health information estimation system can supplement the missing data using past data within a specified date and time range, thereby providing more accurate health information.

[0021] In one embodiment of the health information estimation system, the data construction decision unit determines the data construction based on the temporal analysis results of each data obtained by the data acquisition unit within a specified date and time range.

[0022] According to one embodiment, the health information estimation system determines the data structure based on the temporal analysis results of each data point obtained within a specified date and time range, thereby enabling appropriate data structure determination. Consequently, the health information estimation system can appropriately estimate information related to a person's health. For example, the health information estimation system can estimate not only characteristics obtained from a single moment but also health information based on temporal trends.

[0023] In one embodiment of the health information estimation system, the data construction decision unit determines the data construction based on the temporal analysis results of each data obtained by the first data acquisition unit and the second data acquisition unit within a specified date and time range.

[0024] According to one embodiment, the health information estimation system determines the data structure based on the temporal analysis results of each data point obtained within a specified date and time range, thereby enabling appropriate data structure determination. Consequently, the health information estimation system can appropriately estimate information related to a person's health. For example, by considering the temporal trends of each data point, the health information estimation system can not only estimate features obtained from a single moment of more precise health information, but also estimate health information based on temporal trends.

[0025] In one embodiment of the health information estimation system, the health information estimation unit selects a model from the plurality of health information estimation models that corresponds to the latest data construction to estimate the user's health status.

[0026] According to one embodiment, a health information estimation system estimates a user's health status by selecting a model corresponding to the latest data construction from multiple health information estimation models, thereby enabling estimation based on the latest data. Thus, the health information estimation system can appropriately estimate information related to a person's health. For example, by selecting a model corresponding to the latest data construction, the health information estimation system can accurately estimate the current user's health information.

[0027] In one embodiment of the health information estimation system, the health information estimation unit, in accordance with the data construction based on the temporal analysis results of each data obtained by the data acquisition unit within a specified date and time range, selects a model from the plurality of health information estimation models to estimate the user's health status.

[0028] According to one embodiment, the health information estimation system constructs a corresponding selection model based on the time-series analysis results of each data point to estimate the user's health status, thereby enabling appropriate estimation of the user's health status. Thus, the health information estimation system can appropriately estimate information related to a person's health. For example, the health information estimation system can estimate not only characteristics obtained from a single moment but also health information based on time-series trends.

[0029] In one embodiment of the health information estimation system, the health information estimation unit, in accordance with a plurality of data constructions based on the temporal analysis results of each data obtained by the data acquisition unit within a specified date and time range, selects a model from the plurality of health information estimation models to estimate the user's health status.

[0030] According to one embodiment, the health information estimation system constructs a model corresponding to multiple data points based on the time-series analysis results of each data point to estimate the user's health status, thereby enabling an appropriate estimation of the user's health status. Thus, the health information estimation system can appropriately estimate information related to a person's health. For example, by constructing data based on multiple data points, the health information estimation system can, for instance, estimate health information with higher accuracy based on the temporal trends of each data point.

[0031] In one embodiment of the health information estimation system, the health information estimation unit corrects the health information estimation model based on a correction value calculated corresponding to the data obtained by the data acquisition unit.

[0032] According to one embodiment, the health information estimation system corrects the health information estimation model based on correction values ​​calculated from the acquired data, thereby enabling appropriate estimation of the user's health status. Thus, the health information estimation system can appropriately estimate information related to a person's health. For example, by adjusting coefficients corresponding to the amount of data, the health information estimation system can, for instance, allow the user to refer to health information at the most effective time or at the most opportune time, thereby enabling more accurate estimation of health information.

[0033] In one embodiment of the health information estimation system, the health information estimation unit, in accordance with a plurality of data constructions based on the time-series analysis results of each data obtained by the first data acquisition unit and the second data acquisition unit within a specified date and time range, selects a model from the plurality of health information estimation models to estimate the user's health status.

[0034] According to one embodiment, the health information estimation system constructs a model corresponding to multiple data points based on the time-series analysis results of each data point to estimate the user's health status, thereby enabling an appropriate estimation of the user's health status. Thus, the health information estimation system can appropriately estimate information related to a person's health. For example, by constructing data based on multiple data points, the health information estimation system can, for instance, estimate health information with higher accuracy based on the temporal trends of each data point.

[0035] In one embodiment of the health information estimation system, the health information estimation unit estimates the user's health information at a predetermined period.

[0036] According to one embodiment, the health information estimation system can estimate a user's health information at appropriate timing by estimating the user's health information at predetermined intervals. Thus, the health information estimation system can appropriately estimate information related to a person's health. For example, the health information estimation system can estimate health information at an effective time when the user is most likely to take action.

[0037] In one embodiment of the health information estimation system, the data construction decision unit prompts the user to supplement the external data in order to fill in the gaps in the data arrangement within a specified date and time range.

[0038] According to one embodiment, the health information estimation system prompts the user to supplement external data to compensate for gaps in data arrangement, thereby increasing the likelihood that the missing data will be supplemented and improving the probability of appropriately estimating the user's health information. Thus, the health information estimation system can appropriately estimate information related to a person's health. For example, by prompting the user to supplement external data, the health information estimation system can estimate health information with higher accuracy.

[0039] One embodiment of the information estimation system is characterized by comprising: a data acquisition unit that acquires data detected by sensors configured in the water space; a data construction determination unit that determines a data construction based on a data arrangement or data group formed by associating the data acquired by the data acquisition unit with date and time information; and an information estimation unit that estimates the user's health information or service information provided to the user based on the data acquired by the data acquisition unit, wherein the information estimation unit uses a model corresponding to the data construction from multiple information estimation models.

[0040] According to one embodiment, the information estimation system determines a data structure based on a data arrangement or data group formed by associating data obtained from sensors deployed in a water-using space with date and time information. When estimating a user's health information or service information to be provided to the user based on the data, the system uses a model corresponding to the data structure from multiple information estimation models. This allows the system to estimate the user's health information or service information to be provided to the user in accordance with the acquired data. Therefore, the information estimation system can appropriately estimate information related to a person's health. Furthermore, the information estimation system can determine the data structure based on the acquired data arrangement or data group and use an appropriate model corresponding to it, thereby providing health information or service information without imposing (additional) data input burden on the user.

[0041] In one embodiment of the information estimation system, the sensor is a sensor that detects the composition of urine in excrement discharged into a toilet device, the data acquisition unit acquires urine composition data detected by the sensor, and the data construction determination unit determines the urine data construction based on a urine composition arrangement constructed by associating the urine composition data with date and time information.

[0042] According to one embodiment, an information estimation system acquires urine component data obtained by detecting the composition of urine in excrement discharged into a toilet device using sensors. The system determines the urine data structure based on a urine component arrangement constructed by associating the urine component data with date and time information, thereby enabling the estimation of the user's health information or the provision of service information to the user based on the acquired data. Thus, the information estimation system can appropriately estimate information related to a person's health.

[0043] In one embodiment of the information estimation system, the sensor is a sensor installed in the toilet space to acquire the user's biological data. The data acquisition unit acquires the biological data detected by the sensor, and the data construction determination unit determines the biological data construction based on the biological data arrangement formed by associating the biological data with date and time information.

[0044] According to one embodiment, an information estimation system acquires biological data detected by sensors installed in a toilet space. Based on an arrangement of biological data linked according to date and time information, the system determines the structure of the biological data, thereby enabling the estimation of the user's health information or service information provided to the user based on the acquired data. Thus, the information estimation system can appropriately estimate information related to a person's health.

[0045] The effects of the invention

[0046] According to one method of implementation, information related to a person's health can be appropriately inferred. Attached Figure Description

[0047] Figure 1 This is a diagram illustrating an example of the presumed processing involved in the implementation method.

[0048] Figure 2 This is a diagram illustrating an example of the processing flow in a health information estimation system.

[0049] Figure 3 This is a diagram illustrating an example of data processing.

[0050] Figure 4 This is a diagram illustrating a structural example of a health information estimation system involved in the implementation method.

[0051] Figure 5 This is a block diagram illustrating a structural example of a health information estimation device involved in the implementation method.

[0052] Figure 6 This is a diagram showing an example of a model information storage unit involved in an implementation method.

[0053] Figure 7 This is a diagram illustrating an example of the processing flow in a health information estimation system.

[0054] Figure 8 This is a diagram illustrating an example of data processing.

[0055] Figure 9 This is a diagram illustrating an example of the processing flow in a health information estimation system.

[0056] Figure 10 This is a diagram illustrating an example of data processing.

[0057] Figure 11 This is a diagram illustrating an example of data processing.

[0058] Figure 12 This is a diagram illustrating an example of data processing.

[0059] Figure 13 This is a diagram illustrating an example of data processing.

[0060] Figure 14 This is a diagram illustrating an example of data processing.

[0061] Figure 15 This is a diagram illustrating an example of data processing.

[0062] Figure 16This is a diagram illustrating an example of data processing.

[0063] Figure 17 This is a diagram illustrating an example of data processing.

[0064] Figure 18 This is a diagram illustrating an example of data processing.

[0065] Figure 19 This is a diagram illustrating an example of data processing.

[0066] Figure 20 This is a diagram illustrating an example of data processing.

[0067] Figure 21 This is a diagram illustrating an example of data processing.

[0068] Figure 22 This is a diagram illustrating an example of the processing flow in a health information estimation system.

[0069] Figure 23 This is a diagram illustrating an example of data processing.

[0070] Figure 24 This is a diagram illustrating an example of data processing.

[0071] Figure 25 This is a diagram illustrating an example of data processing. Detailed Implementation

[0072] The health information estimation system and its embodiments disclosed in this application will now be described in detail with reference to the accompanying drawings. Furthermore, this invention is not limited to the embodiments shown below.

[0073] <1. Information Processing Example>

[0074] First, refer to Figure 1 An overview of the information processing performed in the health information estimation system 1 according to the embodiment will be described. Figure 1 This is a diagram illustrating an example of the presumed processing involved in the implementation method. Figure 1 In this paper, the toilet space PS1 is used as an example of a water-using space, but the water-using space is not limited to the toilet space PS1. As long as the health information estimation system 1 can be applied, it can be applied to any water-using space.

[0075] For example, the health information estimation system 1 is not limited to toilet space, but can be applied to any water-using space such as a bathroom (PS2), a space with a washbasin (washing space), or a kitchen (kitchen). Below, without distinguishing between water-using spaces such as toilet space PS1, bathroom space PS2, washbasin space, and kitchen space, it will sometimes be referred to as "water-using space PS".

[0076] The health information estimation system 1 uses data detected by sensor devices 5 configured in the water space PS (also referred to as "water space sensor data") to estimate health information related to the user of the water space PS (also referred to as "user"). For example, health information refers to information representing a person's health-related state (also referred to as "health state"). Here, health state refers to a state related to health, such as, for example, intestinal state, nocturnal urinary frequency, overactive bladder, menstrual cycle, stress, and weakness. Intestinal state also includes various states that can represent a person's intestinal state, such as constipation, diarrhea, and intestinal environment. Furthermore, the above is merely an example; health information is not limited to the above examples, and is not limited to any information related to a person's health.

[0077] First, in use Figure 1 After outlining the processing in the health information estimation system 1, including the transmission and reception of information between the various devices, the system will then use... Figure 2 and Figure 3 The detailed content of each information processing step is explained. Figure 1 The example illustrates the case where the user of the water space PS, i.e., user U1, is used as the presumed subject of health information.

[0078] First, the health information estimation device 100 acquires data detected by the sensor device 5 configured in the water-using space PS (water-using space sensor data) from the sensor device 5 (step S1). For example, the health information estimation device 100 acquires data detected by the sensor device 5 configured in the toilet space PS1, where a toilet device 50, which functions as a so-called toilet bowl, is installed (also referred to as "toilet sensor data"). Figure 1 In this process, the health information estimation device 100 obtains data (toilet sensor data) related to the user U1 detected by the sensor device 5 configured in the toilet space PS1 from the sensor device 5.

[0079] Furthermore, the water-using space sensor data, such as toilet sensor data, can include user biological data, gas data, and fecal data detected by sensor device 5. Additionally, the water-using space sensor data is not limited to toilet sensor data; it can also include data obtained from various water-using spaces (PS). For example, if the water-using space PS considered by the health information estimation system 1 includes a bathroom space PS2, the water-using space sensor data includes data obtained from the bathroom space PS2 (also referred to as "bathroom sensor data"). Similarly, if the water-using space PS considered by the health information estimation system 1 includes a space with a washbasin (washroom space), the water-using space sensor data includes data obtained from the washroom space (also referred to as "washroom sensor data"). Likewise, if the water-using space PS considered by the health information estimation system 1 includes a kitchen space, the water-using space sensor data includes data obtained from the kitchen space (also referred to as "kitchen sensor data"). Details of these points will be described later.

[0080] In addition, Figure 1 For illustrative purposes, the example shown depicts a toilet space PS1 with one toilet device 50. However, multiple toilet devices 50 may be installed in the toilet space PS1, and washbasins, urinals, etc., may also be installed outside of the toilet device 50. Furthermore, the sensor device 5 is not limited to the toilet space PS1, but may also be installed in various water-using spaces PS such as the bathroom space PS2, kitchen space, and washbasin space.

[0081] The health information estimation device 100 obtains data (also referred to as "external data") stored on the server device 2 from the server device 2 (step S2). For example, the external data may include user's diet data, sleep data, exercise data, etc. Figure 1 In this process, the health information estimation device 100 obtains external data about user U1 from the server device 2. For example, the health information estimation device 100 obtains user U1's diet data, sleep data, exercise data, etc., as external data for user U1.

[0082] Furthermore, if the health information estimation device 100 does not use external data in the health information estimation process, it may omit step S2. Additionally, steps S1 and S2 can be performed at any time as long as they are performed before step S3; for example, step S2 may be performed before step S1, or steps S1 and S2 may each be performed multiple times.

[0083] Furthermore, the health information estimation device 100 uses the acquired information to estimate the user's health information (step S3). For example, the health information estimation device 100 uses water space sensor data and external data to estimate the user's health information. Figure 1 In this process, the health information estimation device 100 estimates the health information of user U1 using water space sensor data, etc. For example, the health information estimation device 100 determines the data structure based on the data arrangement formed by associating data obtained about user U1 according to date and time information. Moreover, the health information estimation device 100 selects the model corresponding to the data structure about user U1 from multiple health information estimation models (also simply referred to as "models"), and performs estimation processing to estimate the health status of user U1 using the selected model.

[0084] As described above, the health information estimation system 1, which includes the health information estimation device 100, for example, uses... Figure 2 The processing is performed according to the process shown. Figure 2 This is a diagram illustrating an example of the processing flow in a health information estimation system. (Example:) Figure 2 As shown in the first processing flow PF1, in the health information estimation system 1, the health information estimation device 100 obtains data through a data acquisition unit (e.g., with...). Figure 5 The acquisition unit 131 in the middle acquires water space sensor data (marked as "water space sensor data" in the figure).

[0085] Furthermore, the health information estimation device 100, based on the data arrangement of the acquired water space sensor data, constructs a decision unit (e.g., with...) Figure 5 The data structure of the water space sensor data is determined by the decision unit 132 (corresponding to the decision unit 132 in the middle). Furthermore, the health information estimation device 100 determines the data structure of the water space sensor data by means of the health information estimation unit (e.g., with the decision unit 132 in the middle). Figure 5 The estimation unit 133 in the middle selects a health information estimation model (model) corresponding to the data construction of the determined water space sensor data, and uses the selected model to estimate the user's health information.

[0086] For example, such as Figure 3 As shown in dataset DS1, the health information estimation device 100 determines the data structure of the water space sensor data based on the data arrangement of the acquired water space sensor data, and selects a health information estimation model (model) accordingly based on the determined data structure. Figure 3 This is a diagram illustrating an example of data processing. Figure 3 The dataset DS1 contains data sets that are associated with data obtained at the corresponding date and time, serving as date and time information. Figure 3In the table, as shown in the column labeled "Data Acquired by Sensors," the data obtained are from water space sensor data of the categories corresponding to data #1 to #6. Furthermore, when recording data without distinguishing between datasets such as DS1 and DS2 used in the processing by the health information estimation device 100, it is recorded as dataset DS. For example, dataset DS is stored in storage unit 120 (data storage unit 121, etc.).

[0087] Figure 3 The water space sensor data corresponding to data #1 to #6 can be of any category. For example, data #1 could be data on odorless gases, and data #2 could be data on malodorous gases. Additionally, data #3 could be data on the quantity of feces (also simply referred to as "stool"), data #4 could be data on the shape of feces, and data #5 could be data on the color of feces. Furthermore, data #4 could also be biological data on blood flow. Moreover, the above is just one example; data #1 to #6 can also be data from various categories of water space sensors. For example, data #1 to #6 could be data on different excretion categories, including fecal shape, fecal color, fecal quantity, amount of odorless gases, amount of malodorous gases, urine color, urine quantity, urine flow rate, and urine composition.

[0088] The “xx” placed in the elements (squares) in the third row and subsequent elements of the columns corresponding to data #1 to #6 indicates that the data was acquired at the corresponding time. Figure 3 The data shown is for categories #1, #2, #4, and #5, which were acquired at the time corresponding to "2024 / 3 / 1 8:00". Conversely, it shows data for categories #3 and #6, which were not acquired at the time corresponding to "2024 / 3 / 1 8:00".

[0089] In addition, Figure 3 The image shows data for categories #2 to #5 that were acquired at the time corresponding to "2024 / 3 / 1 12:00". Additionally, in... Figure 3 The image shows data for categories #3 to #6 that were acquired at the time corresponding to "2024 / 3 / 2 9:30". Additionally, in... Figure 3 The data shown is for categories #1, #3 to #6, which were acquired at the time corresponding to "2024 / 3 / 3 18:00".

[0090] Furthermore, the date and time shown can also be a base date and time. In this case, the data can be obtained within a specified date and time range (time period) corresponding to that date and time. For example, the specified date and time range can be a range including several hours before and after the base time (date and time), and can be arbitrarily set. For example, regarding data with "xx" assigned to the row corresponding to "2024 / 3 / 1 8:00", the data in this category can be data obtained within the specified range starting from "2024 / 3 / 1 8:00". For example, if "xx" is assigned to the row corresponding to "2024 / 3 / 1 8:00", the data in this category can be data obtained during a specified period including 8:00 on March 1, 2024 (e.g., 7:00 to 9:00 on March 1, 2024).

[0091] exist Figure 3 In this context, the health information estimation device 100 determines the data structure shown in the 8th column based on the arrangement (data arrangement) of the 7 columns of data from the left (columns 1 to 7) of the dataset DS1, namely the time and sensor-acquired data (#1 to #6). For example, regarding the data corresponding to "2024 / 3 / 1 8:00", since the data arrangement includes data of categories #1, #2, #4, and #5, the health information estimation device 100 determines the data structure corresponding to "2024 / 3 / 1 8:00" as data structure a.

[0092] Furthermore, for example, regarding the data corresponding to "2024 / 3 / 1 12:00", the health information estimation device 100 determines the data structure for the data corresponding to "2024 / 3 / 1 12:00" as data structure b, since the data arrangement includes data from categories #2 to #5. Similarly, the health information estimation device 100 determines the data structure for each piece of data corresponding to "2024 / 3 / 2 9:30" and "2024 / 3 / 3 18:00" as data structures c and d, respectively.

[0093] For example, the health information estimation device 100 can also use overview information (also called a "data structure list") that associates data structure with patterns of data arrangement to determine the data structure corresponding to the data arrangement. For example, the health information estimation device 100 searches the data structure list, determines the data structure a associated with the data arrangement including data of categories #1, #2, #4, and #5, and determines the determined data structure a as the data structure corresponding to "2024 / 3 / 1 8:00". Furthermore, the above is just one example, and the health information estimation device 100 may use any information appropriately to determine the data structure.

[0094] Furthermore, the health information estimation device 100 selects the health information estimation model shown in column 9 based on the data structure shown in column 8. For example, regarding the data corresponding to "2024 / 3 / 1 8:00", since the data structure is data structure a, the health information estimation device 100 selects model A as the health information estimation model corresponding to "2024 / 3 / 1 8:00". Similarly, regarding the data corresponding to "2024 / 3 / 1 12:00", "2024 / 3 / 2 9:30", and "2024 / 3 / 3 18:00", since the data structures are data structures b, c, and d, the health information estimation device 100 selects models B, C, and D respectively as the health information estimation models corresponding to "2024 / 3 / 1 12:00", "2024 / 3 / 2 9:30", and "2024 / 3 / 3 18:00". Furthermore, the health information estimation device 100 uses the selected model to estimate health information.

[0095] As described above, the health information estimation device 100 performs estimation processing using a model selected based on data constructed from data obtained about the user. Thus, the health information estimation system can appropriately estimate information related to a person's health. Furthermore, the health information estimation device 100 can obtain multiple health information estimation models that serve as selection candidates in any manner. For example, the health information estimation device 100 can obtain (receive) multiple health information estimation models from other devices such as a model providing device, and perform estimation processing using the model selected from the obtained multiple health information estimation models. Alternatively, for example, the health information estimation device 100 can generate multiple health information estimation models and perform estimation processing using the model selected from the generated multiple health information estimation models. Further details regarding the models used by the health information estimation device 100 in the estimation processing and examples of the estimation processing will be described later.

[0096] Furthermore, the health information estimation device 100 can also send information based on the estimation process to the user (step S4). For example, the health information estimation device 100 can send the user's health information estimated through the estimation process to the terminal device 10 used by the user. Figure 1In this system, the health information estimation device 100 can send information representing the health status of user U1, estimated through estimation processing, to the terminal device 10 used by user U1. In this case, the terminal device 10, receiving the information from the health information estimation device 100, displays the received information or outputs sound, thereby notifying user U1 of health-related information. Furthermore, the recipients of the notifications related to the estimation results are not limited to user U1; they can also include user U1's family members, other entities related to user U1's health management (medical institutions such as consultation centers), etc.

[0097] <2. Example of a Health Information Estimation System Structure>

[0098] Next, refer to Figure 4 An example of the structure of health information estimation system 1 is illustrated. Figure 4 This is a diagram illustrating a structural example of a health information estimation system involved in the implementation method. Figure 4 This is an example of the structure of a health information estimation system 1. Other structural examples will be described later.

[0099] The health information estimation system 1 includes a health information estimation device 100, a terminal device 10, a sensor device 5, and a server device 2. Alternatively, the health information estimation system 1 may include multiple health information estimation devices 100, multiple terminal devices 10, multiple sensor devices 5, and multiple server devices 2.

[0100] The health information estimation device 100 is an information processing device (computer) that estimates the health information of a user. The health information estimation device 100 can be connected to sensor devices 5, terminal devices 10, and server devices 2 via a predefined network such as the Internet, either wired or wirelessly. The health information estimation device 100 receives various water usage space sensor data, such as toilet sensor data, from sensor device 5. The health information estimation device 100 receives external data from server device 2. The health information estimation device 100 receives data (also called "supplementary data") from terminal device 10 for supplementing missing data. The health information estimation device 100 performs estimation processing using information obtained from at least one of sensor devices 5, terminal devices 10, and server devices 2.

[0101] The health information estimation device 100 determines the data structure based on a data arrangement, which is constructed by associating data obtained from at least one of the sensor device 5, terminal device 10, and server device 2 based on date and time information. The health information estimation device 100 estimates the user's health information based on this data arrangement. For example, the health information estimation device 100 selects a model corresponding to the data structure from multiple health information estimation models to estimate the user's health status.

[0102] Terminal device 10 is an information processing device (equipment) used by a user. Terminal device 10 is implemented, for example, through a smartphone, mobile phone, PDA (Personal Digital Assistant), tablet computer, or laptop PC (Personal Computer). Figure 1 In the example shown, terminal device 10 is a smartphone used by the user.

[0103] Terminal device 10 sends data (information) to health information estimation device 100 in response to user operations. Additionally, terminal device 10 outputs information received from health information estimation device 100. Terminal device 10 can communicatively connect to server device 2, health information estimation device 100, etc., via a specified network such as the Internet, and send and receive information between them. In response to requests from health information estimation device 100, terminal device 10 sends supplementary data to supplement missing data. Terminal device 10 displays information received from health information estimation device 100. Furthermore, terminal device 10 can also output sound from information received from health information estimation device 100.

[0104] Sensor device 5 performs detection related to various water-using space sensor data, such as toilet sensor data. Sensor device 5 is disposed in the water-using space PS and performs detection within the water-using space PS. For example, sensor device 5 is disposed in toilet space PS1 where toilet device 50 is installed. Toilet space PS1 where toilet device 50 is installed is not limited to a user's residence, but can also be a toilet (public toilet) installed in a building or other facility. Sensor device 5 is not limited to toilet space PS1, and can also be disposed in any water-using space PS such as bathroom space PS2, kitchen space, or washroom space.

[0105] For example, the sensor device 5 installed in the toilet space PS1 includes a gas sensor. In this case, the sensor device 5 detects gas data using the gas sensor, and the gas data can be included in the water space sensor data. The gas sensor of the sensor device 5 detects gases released from the user's body to the outside in the toilet space PS1. For example, the gas sensor of the sensor device 5 detects the user's excrement gases (such as flatulence), gases released from the user's excrement, etc. For example, the sensor device 5 can have a gas sensor (odorless gas sensor) that detects odorless gases composed of hydrogen, methane, carbon dioxide, etc. Alternatively, the sensor device 5 can also have a gas sensor (malodorous gas sensor) that detects malodorous gases composed of hydrogen sulfide, methanethiol, etc.

[0106] Furthermore, the sensor device 5 is not limited to the aforementioned devices, and may also include a gas sensor that detects any gas. For example, the sensor device 5 may also include a gas sensor (health-type gas sensor) that detects gases that indicate high health levels (also called "health-type gases") produced by fermentation in the intestines. For example, the sensor device 5 may also include a gas sensor (health-type gas sensor) that detects at least one of the following health-type gases: hydrogen, carbon dioxide, acetic acid, methane, ethanol, and water.

[0107] For example, sensor device 5 may also have a gas sensor (odor gas sensor) that detects gases (also called "odorous gases") that indicate low health due to putrefaction in the intestines. For example, sensor device 5 may also have a gas sensor (odor gas sensor) that detects at least one of odorous gases such as ammonia, trimethylamine, hydrogen sulfide, methanethiol, indole, and skatole.

[0108] Additionally, the sensor device 5 installed in the toilet space PS1 can also be a stool sensor. In this case, the sensor device 5 detects stool data using a gas sensor, and the stool data can be included in the water space sensor data. For example, the stool sensor of the sensor device 5 is a sensor that captures images of stool using a line sensor, a camera (two-dimensional image sensor), or the like. In this case, the stool sensor of the sensor device 5 captures images of the user's excreted feces (sometimes simply referred to as "stool"), generating an image containing the user's stool. Furthermore, the stool sensor of the sensor device 5 can use any sensor as long as it can detect at least one of the stool characteristics, such as color, shape, and quantity. Furthermore, the determination of the user's stool characteristics can be performed by the sensor device 5 or by the health information estimation device 100. Additionally, the stool sensor can also detect the user's urine. For example, the stool sensor can also detect the color and quantity of the user's urine. In this case, the sensor device 5 detects urine data using a gas sensor, and the urine data can be included in the water space sensor data.

[0109] Furthermore, the above is merely one example, and the sensor device 5 may also include various other sensors besides those described above. For example, the sensor device 5 configured in the toilet space PS1 may include a sensor that detects when someone sits on the toilet seat (seat detection sensor). Additionally, the sensor device 5 may also include a sensor that detects the biological data of the user in the water-using space PS. For example, the sensor device 5 may include a sensor that detects the user's blood flow, heartbeat (pulse), respiration, body temperature, brain waves, muscles (electromyography), and other biological data. For example, in the case of detecting the user's blood flow in the toilet space PS1, the sensor device 5 may include a blood flow sensor installed on the toilet seat or similar component of the toilet device 50. In this case, the sensor device 5 detects blood flow data using the blood flow sensor, and the blood flow data can be included in the water-using space sensor data. As described above, the water-using space sensor data may include various biological data such as blood flow data, heartbeat data, respiration data, body temperature data, brain wave data, and muscle data. Furthermore, the sensor device 5 may also include a human body detection sensor that detects whether anyone is in the water-using space PS.

[0110] Sensor device 5 has communication capabilities, enabling it to send and receive information with health information estimation device 100 via a specified network such as the Internet. Sensor device 5 may also have an information parsing unit. Sensor device 5 transmits data (information) collected through detection to health information estimation device 100. Sensor device 5 transmits various water-using space sensor data, such as toilet sensor data, to health information estimation device 100.

[0111] Server device 2 stores the data (external data) used by health information estimation device 100 in processing and provides the external data to the data storage device (service providing device) of health information estimation device 100. Server device 2 may be a cloud containing multiple server devices, etc. For example, server device 2 can send and receive information with health information estimation device 100, terminal device 10, sensor device 5, etc. via a defined network such as the Internet. Server device 2 sends external data to health information estimation device 100. In response to a request from health information estimation device 100, server device 2 sends external data corresponding to the request from health information estimation device 100 to health information estimation device 100.

[0112] Furthermore, the above is merely an example; the health information estimation system 1 can employ any device structure as long as the desired processing can be achieved. For example, the health information estimation system 1 may also include various sensors for collecting water space sensor data, such as toilet sensor data. For instance, as an entry detection sensor for collecting information on users' entry and exit from the toilet space, the health information estimation system 1 may include a door sensor for detecting the opening and closing of the toilet space door. In this case, the door sensor communicates with the sensor device 5, sending information related to the user's entry and exit to the sensor device 5. As described above, the system structure is merely an example; the health information estimation system 1 can be any system structure as long as the desired processing can be achieved.

[0113] Additionally, the health information estimation device 100 can also receive data detected by the sensor device 5 via the terminal device 10. For example, the health information estimation device 100 can also receive data about the user detected by the sensor device 5 via the user's terminal device 10. In this case, the terminal device 10 can communicatively connect to the sensor device 5 via a specified network such as the Internet, and send and receive information with the sensor device 5. Furthermore, the communication method can be arbitrary; the terminal device 10 can communicate with the sensor device 5 via specified wireless communication functions such as Bluetooth (registered trademark) or Wi-Fi (Wireless Fidelity) (registered trademark).

[0114] Terminal device 10 receives water usage space sensor data, such as toilet sensor data, detected by sensor device 5 regarding the user of the terminal device 10. Furthermore, terminal device 10 transmits the water usage space sensor data received from sensor device 5 to health information estimation device 100. In this case, terminal device 10 transmits the water usage space sensor data received from sensor device 5, along with information identifying the user of the terminal device 10 (user ID, etc.), to health information estimation device 100. The health information estimation device 100, having received the water usage space sensor data and user identification information (user ID, etc.) from terminal device 10, registers the water usage space sensor data and user identification information (user ID, etc.) in association with storage unit 120 (data storage unit 121, etc.).

[0115] <3. Functional Structure of Health Information Estimation Device>

[0116] Below, refer to Figure 5 The functional structure of the health information estimation device 100 is explained. Figure 5 This is a block diagram illustrating a structural example of a health information estimation device involved in the implementation method.

[0117] like Figure 5 As shown, the health information estimation device 100 includes a communication unit 110, a storage unit 120, and a control unit 130. Furthermore, the health information estimation device 100 may also include an input unit (e.g., a keyboard or mouse) for receiving various operations from an administrator or other personnel, and a display unit (e.g., a liquid crystal display) for displaying various information.

[0118] The communication unit 110 is implemented, for example, through a communication circuit. The communication unit 110 is connected to a specified network such as the Internet via wired or wireless means, and performs information transmission and reception with external information processing devices. For example, the communication unit 110 performs information transmission and reception with other devices with communication functions, such as server device 2, terminal device 10, and sensor device 5, via a specified network such as the Internet.

[0119] The storage unit 120 is implemented, for example, by a semiconductor memory element such as RAM (Random Access Memory) or flash memory, or a storage device such as a hard disk or optical disk. For example, the storage unit 120 is a computer-readable recording medium that records non-temporary data such as data used by a health information estimation program.

[0120] like Figure 5 As shown, the storage unit 120 in the embodiment includes a data storage unit 121, a model information storage unit 122, and a user information storage unit 123.

[0121] The data storage unit 121 in this embodiment stores various data used by the health information estimation device 100 during processing. The data storage unit 121 stores various data used for estimating the health information of users utilizing the water space. The data storage unit 121 stores various data obtained regarding users utilizing the water space.

[0122] The data storage unit 121 stores water space sensor data, such as toilet sensor data. The data storage unit 121 stores each piece of data detected by the sensor device 5 along with the date and time information of the date the data was acquired, as water space sensor data. For example, the data storage unit 121 stores toilet sensor data acquired regarding the toilet space PS1. For example, the data storage unit 121 stores toilet sensor data including gas data, feces data, urine data, and biological data detected in the toilet space PS1. For example, the data storage unit 121 stores gas data including one piece of data from various gases such as odorless gases and malodorous gases, and feces data including at least one piece of data from the quantity, shape, and color of feces.

[0123] As described above, the data storage unit 121 may also contain data such as the shape, color, amount, amount of odorless gas, amount of foul-smelling gas, color, amount, flow rate, and composition of urine. For example, various categories of data, such as the shape, color, amount, amount of odorless gas, amount of foul-smelling gas, color, amount, flow rate, and composition of urine, can be generated by the sensor device 5 or by the health information estimation device 100 through analysis processing. For example, data such as the shape, color, amount, color, amount, flow rate, and composition of urine can be generated based on images obtained by photographing stool or urine. Regarding this, an example of stool shape will be explained. The health information estimation device 100 can use AI (artificial intelligence) related technologies to determine the shape of stool. For example, the health information estimation device 100 can use an AI model (machine learning model) generated through machine learning, i.e., a shape determination model, to determine the shape of stool.

[0124] In this case, the shape determination model is learned beforehand using teacher data representing classification judgments. Multiple combinations of images containing feces (also called "feces images") and labels (correct answer information) representing the shape (one of type 1 to type 7) of the blocks (feces) contained in that feces image are included in the teacher data. For example, types 1 to 7 could be seven categories of feces shape based on the Bristol feces classification system. For example, types 1 to 7 correspond to granular, lumpy, cracked, banana-shaped, soft, muddy, and watery, respectively.

[0125] For example, a shape determination model is a model that takes a stool image as input and outputs information representing the shape of the blocks (stool) contained in the input stool image. For example, a shape determination model can be learned by outputting information about the labels (shapes of stool) corresponding to the input stool image when given a stool image. The learning of the shape determination model is performed appropriately using various methods related to so-called teacher-guided learning. In this case, the shape determination model is stored in storage unit 120, and the health information estimation device 100 can use the shape determination model stored in storage unit 120 to determine the shape of stool. For example, the health information estimation device 100 can perform learning processing to generate a shape determination model.

[0126] Furthermore, the above is merely an example; the health information estimation device 100 can also appropriately use various information to determine the shape of stool. Additionally, the seven categories mentioned above—granular, lumpy, cracked, banana-shaped, soft, muddy, and watery—are just examples of shape; the health information estimation device 100 can also determine other shapes, and may classify them as six categories or less. Furthermore, the determination (analysis) of stool shape can also be performed by the sensor device 5. In this case, the sensor device 5 can use a shape determination model, etc., to determine (analyze) the shape of stool, and the health information estimation device 100 obtains data representing the shape of stool from the sensor device 5 as water space sensor data. Furthermore, data on categories other than stool shape, such as stool color, stool volume, odorless gas (amount), foul-smelling gas (amount), urine color, urine volume, urine flow rate, and urine composition, can also be generated using an AI model in the same way as the stool shape data.

[0127] In addition to toilet sensor data, data storage unit 121 also stores various water-using space sensor data. For example, data storage unit 121 stores bathroom sensor data acquired regarding the bathroom space PS2. For example, data storage unit 121 stores bathroom sensor data including data on organisms detected in the bathroom space PS2. For example, data storage unit 121 stores washroom sensor data acquired regarding the washroom space. For example, data storage unit 121 stores washroom sensor data including data on organisms detected in the washroom. For example, data storage unit 121 stores kitchen sensor data acquired regarding the kitchen space. For example, data storage unit 121 stores kitchen sensor data including data on organisms detected in the kitchen space.

[0128] For example, the data storage unit 121 can store the analysis results obtained by analyzing the data detected by the sensor device 5 as water space sensor data. For example, the data storage unit 121 can store data such as toilet sensor data and water space sensor data obtained by analyzing the sensor information detected by the sensor device 5.

[0129] In addition, the data storage unit 121 stores external data. For example, the data storage unit 121 stores external data (external information) obtained from the server device 2. The data storage unit 121 stores various data related to the user's actions. The data storage unit 121 stores motion data (motion information) related to motion. For example, the data storage unit 121 stores various motion data related to the user's motion. For example, the data storage unit 121 stores motion data representing the content of the user's motion. For example, the data storage unit 121 stores motion data representing the amount of motion of the user. For example, the data storage unit 121 stores motion data representing the time of motion of the user.

[0130] Data storage unit 121 stores diet-related data (dietary information). For example, data storage unit 121 stores diet data indicating the user's diet content. For example, data storage unit 121 stores diet data indicating the user's diet quantity. For example, data storage unit 121 stores diet data indicating the user's diet time. Data storage unit 121 stores sleep-related data (sleep information). For example, data storage unit 121 stores sleep data indicating the user's sleep content. For example, data storage unit 121 stores sleep data indicating the user's sleep volume. For example, data storage unit 121 stores sleep data indicating sleep duration.

[0131] Furthermore, the data storage unit 121 is not limited to the information described above, and can store various types of information depending on the purpose. For example, when there are multiple users, the data storage unit 121 may associate user identification information (e.g., user ID) with that user's information and store it. For example, the data storage unit 121 may store information related to the cleaning of the sensor device 5. For example, the data storage unit 121 may store data indicating the number of cleaning operations. In addition, the data storage unit 121 can be divided into: a water space sensor data storage unit, which stores water space sensor data based on detection performed by the sensor device 5; and an external data storage unit, which stores external data.

[0132] The model information storage unit 122 in the embodiment stores information related to the model. For example, the model information storage unit 122 stores information about multiple models (health information estimation models) used to estimate the user's health information. Figure 6 This is a diagram illustrating an example of the model information storage unit involved in the implementation method. Figure 6 The model information storage unit 122 shown contains items such as "model ID", "health information estimation model", "data construction" and "model data".

[0133] "Model ID" represents the identification information used to identify the model. "Health Information Presumption Model" represents the content of the model identified by the Model ID. The information in the "Health Information Presumption Model" item shows the model content, which may include various information related to the model's content, such as its name, purpose, and category.

[0134] "Data Construction" refers to the data construction associated with the health information presumption model. "Data Construction" indicates the data construction used by the corresponding health information presumption model. The information displayed in the "Data Construction" section may include various information related to the content of that data construction, such as the data arrangement (pattern) corresponding to that data construction.

[0135] "Model data" refers to the data of the model. Figure 6 The example shown is that conceptual information such as "MDT1" is stored in "model data", but in reality, it contains various information that constitutes the model, such as network-related information and functions.

[0136] exist Figure 6 In the example shown, the model identified by model ID "M1", i.e., model A, represents the health information estimation model used when the data structure is data structure a. Furthermore, the model data representation for model A is model data MDT1.

[0137] For example, model A could be a model that constructs the input data corresponding to data 'a' and outputs the user's health information inferred from that data. In this case, model A could be a machine learning model (also known as an "AI model") trained (learned) through machine learning.

[0138] For example, Model A could be an AI model that, given data structure a of data obtained about a user, outputs inferred health information about that user based on the input of that data. For example, Model A could be an AI model that, based on the input of data corresponding to data structure a, outputs a score (value) representing a certain health state (e.g., intestinal state, nocturnal urination, overactive bladder, menstrual cycle, stress, weakness, etc.). Furthermore, as mentioned above, regarding intestinal state, refined states such as constipation, diarrhea, and the intestinal environment can be inferred. In this case, for example, if Model A infers intestinal state, Model A could be an AI model that, based on the input of data corresponding to data structure a, outputs a score (value) representing the good or bad of at least one of constipation, diarrhea, or the intestinal environment.

[0139] Alternatively, Model A can also be an AI model that constructs the input data corresponding to data 'a' and outputs information representing the classification of the user's health status corresponding to that input. For example, Model A can be a classifier (recognizer) that constructs the input data corresponding to data 'a' and outputs information representing whether the user's health status corresponding to that input is (classified as) good (health status), intestinal status, nocturnal urination frequency, overactive bladder, menstrual cycle, stress, weakness, etc. In this case, it could mean: intestinal status is poor intestinal status; nocturnal urination frequency is nocturnal urination frequency; overactive bladder is overactive bladder; menstrual cycle is menstrual cycle disorder, etc.; different from the expected state; stress is a state of excessive stress; weakness is a state of weakness. As mentioned above, Model A can be a function that generates new information based on data, like an AI model. The function mentioned here, as long as it generates new information based on data, is not limited to AI models and includes various things such as programs and algorithms.

[0140] Furthermore, model A is not limited to functions like those described above; it can be any model capable of inferring the user's health information. For example, model A could represent information about the health status of a user who has obtained data corresponding to data construction a. In this case, model A could represent information about the type of health status (also called "type information") to which the user who has obtained data corresponding to data construction a is classified. For example, model A could represent type information such as whether the user who has obtained data corresponding to data construction a is classified as having good health, good bowel function, frequent urination at night, overactive bladder, menstrual cycle, stress, or weakness.

[0141] Furthermore, the model identified by model ID "M2", namely model B, represents a health information estimation model used when the data structure is data structure b. Additionally, the model data representation of model B is model data MDT2. Moreover, similar to model A, model B can be a function such as an AI model, or information representing the user's health status such as type information.

[0142] Furthermore, the model identified by model ID "M3", namely model C, represents a health information estimation model used when the data structure is data structure c. Additionally, the model data representation of model C is model data MDT3. Moreover, similar to model A, model C can be a function such as an AI model, or information representing the user's health status such as type information.

[0143] Furthermore, the model identified by model ID "M4", namely model D, indicates a health information estimation model used when the data structure is data structure d. Additionally, the model data representation of model D is model data MDT4. Moreover, similar to model A, model D can be a function such as an AI model, or information representing the user's health status such as type information.

[0144] Furthermore, the model information storage unit 122 is not limited to the above; it can also store various types of information depending on the purpose. For example, models A to D described above are just one example; other models can also be stored in the model information storage unit 122, and more than five models can be stored.

[0145] The user information storage unit 123 in the embodiment stores user information. For example, the user information storage unit 123 stores various user information related to users of the water-using space. For example, the user information storage unit 123 stores information that identifies the user (user ID) in association with the user's information.

[0146] The user information storage unit 123 stores various information related to the attributes of users identified by their user IDs. For example, the user information storage unit 123 stores various user attribute information such as demographic attributes like age and gender, and psychometric attributes like lifestyle and hobbies. Attribute information may include, for example, weight, height, mental state, and medical history.

[0147] Furthermore, the user information storage unit 123 is not limited to the above, and can also store various types of information depending on the purpose. For example, the user information storage unit 123 may store user information collected through user input. In this case, the user information storage unit 123 may store information representing data supplemented by the user (supplementary data).

[0148] Furthermore, the above is merely one example; the storage unit 120 stores various types of information used in the processing. In addition to the above, the storage unit 120 also stores various types of information used in the presumption processing of health information. For example, the storage unit 120 stores a summary of information (a data structure list) that associates data structure with patterns of data arrangement.

[0149] The control unit 130 is implemented, for example, by executing a program (such as the health information estimation program disclosed herein) stored internally in the health information estimation device 100 using RAM or the like as a working area by a CPU (Central Processing Unit) or GPU (Graphics Processing Unit). Alternatively, the control unit 130 is implemented, for example, by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

[0150] The control unit 130 includes an acquisition unit 131, a decision unit 132, an estimation unit 133, and a transmission unit 134, and performs or executes the information processing functions or roles described below. Furthermore, the internal structure of the control unit 130 is not limited to... Figure 4 The structure shown can be any other structure, as long as it is the structure for information processing described later.

[0151] Acquisition unit 131 acquires various types of information. Acquisition unit 131 acquires various types of information from storage unit 120. Acquisition unit 131 receives various types of information from various computers such as sensor device 5 or terminal device 10. Acquisition unit 131 acquires various data (information) used in estimation processing, such as water space sensor data, external data, and supplementary data. For example, acquisition unit 131 acquires water space sensor data from data storage unit 121. For example, acquisition unit 131 acquires external data from data storage unit 121. For example, acquisition unit 131 acquires model information from model information storage unit 122. For example, acquisition unit 131 acquires user information from user information storage unit 123.

[0152] The acquisition unit 131 functions as a data acquisition unit that acquires data (such as water space sensor data) detected by the sensor device 5 arranged in the water-using space PS. The acquisition unit 131 functions as a first data acquisition unit that acquires data (such as toilet sensor data) detected by the sensor device 5 arranged in the toilet space. The acquisition unit 131 also functions as a second data acquisition unit that acquires external data.

[0153] The acquisition unit 131 acquires water space sensor data from the sensor devices 5 of each water space PS by receiving water space sensor data from the sensor devices 5 of each water space PS. The acquisition unit 131 stores (registers) the water space sensor data acquired from the sensor devices 5 of each water space PS in the data storage unit 121. The acquisition unit 131 acquires toilet sensor data from the sensor devices 5 of the toilet space PS1. The acquisition unit 131 acquires toilet sensor data from the sensor devices 5 of the toilet space PS1, including gas data, feces data, urine data, biological data, etc. For example, the acquisition unit 131 acquires gas data containing one data point from various gases such as odorless gas and malodorous gas, and feces data containing at least one data point from the quantity, shape, and color of feces from the sensor devices 5 of the toilet space PS1.

[0154] The acquisition unit 131 acquires bathroom sensor data from the sensor device 5 of the PS2 in the bathroom space. The acquisition unit 131 acquires bathroom sensor data, including biological data, from the sensor device 5 of the PS2 in the bathroom space. The acquisition unit 131 acquires washroom sensor data from the sensor device 5 of the sensor device 5 in the washroom space. The acquisition unit 131 acquires washroom sensor data, including biological data, from the sensor device 5 in the sensor device 5 in the sensor device 5 in the kitchen space. The acquisition unit 131 acquires kitchen sensor data, including biological data, from the sensor device 5 in the sensor device 5 in the kitchen space.

[0155] The decision unit 132 performs decision processing based on various information. For example, the decision unit 132 performs decision processing using various information stored in the storage unit 120. For example, the decision unit 132 performs decision processing using various information obtained by the acquisition unit 131.

[0156] The decision unit 132 performs decision processing using information received from various computers such as the server device 2, sensor device 5, and terminal device 10. For example, the decision unit 132 performs decision processing based on a water usage log that includes sensor information detected by the sensor device 5. For example, the decision unit 132 performs decision processing based on external data obtained from the server device 2.

[0157] The decision unit 132 functions as a data structure determination unit that determines the data structure based on a data arrangement formed by associating data acquired by the acquisition unit 131 with date and time information. The decision unit 132 determines the data structure based on a data arrangement formed by associating at least one type of data, including water space sensor data and external data, with date and time information.

[0158] The decision unit 132 determines the data structure based on the data arrangement formed by associating data obtained by the acquisition unit 131 with personal information or date information. The decision unit 132 determines the data structure based on a combination of at least one or more data in the data arrangement.

[0159] The decision unit 132 supplements the missing data arrangement in the latest data arrangement using past data arrangements within a specified date and time range, and determines the data structure based on the supplemented data arrangement. The decision unit 132 determines the data structure based on the time-series analysis results of each data acquired by the acquisition unit 131 within the specified date and time range. In order to supplement the missing data arrangement within the specified date and time range, the decision unit 132 prompts the user to supplement external data. For example, the decision unit 132 prompts the user to supplement external data by having the sending unit 134 send a message requesting the user to supplement external data.

[0160] The estimation unit 133 performs estimation processing for various types of information. For example, the estimation unit 133 performs estimation processing using various types of information stored in the storage unit 120. For example, the estimation unit 133 performs estimation processing using various types of information obtained by the acquisition unit 131. For example, the estimation unit 133 performs estimation processing using various types of information obtained by the decision unit 132.

[0161] The estimation unit 133 performs estimation processing using information received from various computers such as server device 2, sensor device 5, and terminal device 10. For example, the estimation unit 133 performs estimation processing based on a water usage space log that includes sensor information detected by sensor device 5. For example, the estimation unit 133 performs estimation processing based on external data obtained from server device 2.

[0162] The estimation unit 133 functions as a health information estimation unit that estimates the user's health information based on data arrangement. The estimation unit 133 selects the model corresponding to the data construction from multiple health information estimation models to estimate the user's health status.

[0163] For example, if the health information estimation model is a function such as an AI model, the estimation unit 133 uses the information output by the model by inputting data into the selected model to estimate the user's health status. For example, if the selected model is an AI model that outputs a score (value) representing a health status based on the input data, the estimation unit 133 estimates whether the user of the data is in that health status based on the score output by the model by inputting data into the selected model.

[0164] In this case, for example, if the selected model is an AI model that outputs a score (value) representing nighttime urination frequency based on data input, and the score output by the model after inputting data into the selected model is above a predetermined threshold, the estimation unit 133 estimates that the user of the data has nighttime urination frequency. Alternatively, for example, if the selected model is an AI model that outputs a score (value) representing nighttime urination frequency based on data input, and the score output by the model after inputting data into the selected model is below a predetermined threshold, the estimation unit 133 estimates that the user of the data does not have nighttime urination frequency.

[0165] Furthermore, for example, if the selected model is an AI model that outputs information representing a classification of the user's health status based on the input data, the estimation unit 133 estimates the user's health status based on the information output by the model when data is input to the selected model. For example, the estimation unit 133 estimates the user's health status as good, intestinal condition, nocturnal urination, overactive bladder, menstrual cycle, stress, frailty, etc., based on the information output by the model when data is input to the selected model.

[0166] Furthermore, the estimation unit 133 can also perform learning processing on the AI ​​model used in the estimation process. For example, the estimation unit 133 uses learning data (teacher data, etc.) to learn the AI ​​model. This learning data includes multiple learning data sets that combine data with labels (correct answer information) representing the health status of the user corresponding to that data. When the estimation unit 133 performs learning processing, the learning data used in the learning processing can also be stored in the storage unit 120.

[0167] For example, when an AI model is used to estimate whether a user is in a certain health state, it takes data as input and outputs information representing the user's health state (the value (score)) corresponding to the input data. Similarly, when an AI model is used to classify and estimate a user's health state, it takes data as input and outputs information representing the user's health state (the classification result) corresponding to the input data.

[0168] For example, an AI model learns by outputting information about labels corresponding to the input data, given the input data. The learning of the AI ​​model is carried out using various methods related to so-called teacher-guided learning. Furthermore, the learning of the AI ​​model is not limited to the methods described above; any learning method can be used as long as the desired AI model can be generated. Additionally, if the health information estimation device 100 does not perform learning processing and instead obtains the AI ​​model from other devices (such as a model providing device), the estimation unit 133 may also not perform learning processing. As described above, the health information estimation device 100 can generate the model used for estimation processing, or it can use the model used for estimation processing obtained from other devices such as a model providing device.

[0169] Furthermore, when the health information estimation model is information representing the user's health status, such as type information, the estimation unit 133 estimates the user's health status based on the health status shown by the selected model. For example, when the health information estimation model is type information representing the type (category) of the user's health status, the estimation unit 133 estimates the health status shown by the selected model as the user's health status.

[0170] In such cases, for example, if the health information estimation model, i.e., the type information, indicates that the user's health status is good, then the estimation unit 133 estimates the user's health status as good. For example, if the health information estimation model, i.e., the type information, indicates that the user's health status is poor intestinal health, then the estimation unit 133 estimates the user's health status as poor intestinal health. For example, if the health information estimation model, i.e., the type information, indicates that the user's health status is under stress and has overactive bladder, then the estimation unit 133 estimates the user's health status as under stress and has overactive bladder.

[0171] The estimation unit 133 selects the model corresponding to the latest data construction from multiple health information estimation models to estimate the user's health status. Corresponding to the data construction based on the time-series analysis results of each data obtained by the acquisition unit 131 within a specified date and time range, the estimation unit 133 selects a model from multiple health information estimation models to estimate the user's health status.

[0172] The estimation unit 133 constructs multiple data based on the time-series analysis results of each data acquired by the acquisition unit 131 within a specified date and time range, and selects a model from multiple health information estimation models to estimate the user's health status. The estimation unit 133 corrects the health information estimation model based on correction values ​​calculated corresponding to the data acquired by the acquisition unit 131. The estimation unit 133 estimates the user's health information at a predetermined period. For example, the estimation unit 133 estimates the user's health information at a predetermined period such as 12:00 noon every day, Monday every week, or the 1st of every month. Alternatively, the estimation unit 133 may estimate the user's health information at a predetermined period such as when new data is acquired.

[0173] Alternatively, the estimation unit 133 may also function as an information generation unit that generates various types of information notified by the sending unit 134. In this case, the estimation unit 133 generates various types of information using information obtained from other devices or information stored in the storage unit 120. The estimation unit 133 generates information using information obtained by the acquisition unit 131. The estimation unit 133 generates information using the estimation result.

[0174] The estimation unit 133 appropriately uses various technologies to generate various information, such as screens (image information), to be provided to external information processing devices. The estimation unit 133 generates screens (image information) to be provided to the terminal device 10. For example, the estimation unit 133 generates screens (image information) to be provided to the terminal device 10 based on information stored in the storage unit 120.

[0175] As long as the estimation unit 133 can generate a screen (image information) to be provided to an external information processing device, it can generate a screen (image information) through arbitrary processing. For example, the estimation unit 133 appropriately uses various technologies related to image generation or image processing to generate a screen (image information) to be provided to the terminal device 10. For example, the estimation unit 133 appropriately uses various technologies such as Java (registered trademark) to generate a screen (image information) to be provided to the terminal device 10.

[0176] The sending unit 134 sends information to external information processing devices. For example, the sending unit 134 sends various types of information to computers such as server device 2, sensor device 5, and terminal device 10. The sending unit 134 notifies the user of information related to the user's health information. The sending unit 134 notifies the user by sending information estimated by the estimation unit 133 to the terminal device 10. The sending unit 134 sends information generated by the estimation unit 133 to the terminal device 10. In order to supplement missing data in accordance with the instructions from the decision unit 132, the sending unit 134 prompts the user to supplement external data. The sending unit 134 sends a message requesting the user to supplement external data to the terminal device 10 used by the user.

[0177] <4. Other processing examples>

[0178] In addition, Figures 2-3 The processing shown is merely an example; the health information estimation system 1 is not limited to the above example and can perform various processing methods. Several examples are given to illustrate this point. Furthermore, explanations of points that are the same as those described above are appropriately omitted.

[0179] For example, the health information estimation system 1 is not limited to Figure 2 Alternatively, it can be done through Figure 7 The processing flow shown is executed. Figure 7 This is a diagram illustrating an example of the processing flow in a health information estimation system. (Example:) Figure 7 As shown in the second processing flow PF2, in the health information estimation system 1, the health information estimation device 100 obtains data through the first data acquisition unit (e.g., with...). Figure 5 The data acquisition unit 131 in the middle acquires water space sensor data. Additionally, the health information estimation device 100 acquires data through the first data acquisition unit (e.g., with...). Figure 5 (corresponding to the acquisition part 131 in the middle) to acquire external data.

[0180] Furthermore, the health information estimation device 100, based on the data arrangement of the acquired water space sensor data and external data, constructs a decision unit (e.g., with...) Figure 5 The decision unit 132 in the middle determines the data structure of the water space sensor data and external data. Furthermore, the health information estimation device 100 uses a health information estimation unit (e.g., with...) Figure 5 The estimation unit 133 in the middle selects a health information estimation model (model) corresponding to the data construction of the determined water space sensor data and external data, and uses the selected model to estimate the user's health information.

[0181] For example, Figure 8 As shown in dataset DS2, the health information estimation device 100 determines the data structure of the water space sensor data and external data based on the data arrangement of the acquired water space sensor data and external data, and selects a health information estimation model (model) accordingly based on the determined data structure. Figure 8 This is a diagram illustrating an example of data processing. Figure 8 The DS2 dataset contains data sets that associate date and time information, i.e., time, with the data obtained at the corresponding date and time. Figure 8The following situations are illustrated: as shown in the column labeled "Sensor Acquired Data," water space sensor data of the category corresponding to data #1 to #3 is acquired; as shown in the column labeled "External Data," external data of the category corresponding to data #1 to #3 is acquired. Hereinafter, water space sensor data #1 to #3 will sometimes be recorded as water space data #1 to #3, and external data #1 to #3 will sometimes be recorded as external data #1 to #3.

[0182] Figure 8 The water space sensor data corresponding to the categories of water space data #1 to #3 can be data of any category. For example, water space data #1 can be gas data of odorless gases, and water space data #2 can be gas data of malodorous gases. Additionally, water space data #3 can be urine data, for example. Furthermore, the above is just one example; water space data #1 to #3 can also be water space sensor data of various categories.

[0183] Figure 8 The external data corresponding to the categories of external data #1 to #3 can be data of any category. For example, external data #1 could be dietary data, external data #2 could be sleep data, and external data #3 could be exercise data. Furthermore, the above is just one example; external data #1 to #3 can also be external data of various categories.

[0184] exist Figure 8 The data shown is for the categories of Water Use Space Data #1, Water Use Space Data #2, External Data #1, and External Data #2, obtained at the time corresponding to "2024 / 3 / 1 8:00". However, it also shows that at the time corresponding to "2024 / 3 / 1 8:00", data for categories of Water Use Space Data #3 and External Data #3 were not obtained.

[0185] In addition, Figure 8 The document shows data categorized as water usage spatial data #2, #3, and external data #1, #2, acquired at the time corresponding to "2024 / 3 / 1 12:00". Additionally, in... Figure 8 The data shown is for categories #3 (water space data) and #1 to #3 (external data) obtained at the time corresponding to "2024 / 3 / 2 9:30". Additionally, in Figure 8 The data shown is for the categories of water use spatial data #1, #3, and external data #1 to #3, obtained at the time corresponding to "2024 / 3 / 3 18:00".

[0186] exist Figure 8In this process, the health information estimation device 100 determines the data structure shown in the 8th column based on the arrangement (data arrangement) of the 7 columns of data from left to right in the dataset DS2, namely, time, data acquired by the first sensor (#1 to #3), and external data (#1 to #3). For example, regarding the data corresponding to "2024 / 3 / 1 8:00", since the data arrangement includes data of categories such as water space data #1, #2, and external data #1, #2, the health information estimation device 100 determines the data structure corresponding to "2024 / 3 / 1 8:00" as data structure a.

[0187] Furthermore, for example, regarding the data corresponding to "2024 / 3 / 1 12:00", the health information estimation device 100 determines data structure b for the data corresponding to "2024 / 3 / 1 12:00" because the data arrangement includes data of categories such as water usage space data #2, #3, and external data #1, #2. Similarly, the health information estimation device 100 determines data structures c and d for the data corresponding to "2024 / 3 / 2 9:30" and "2024 / 3 / 3 18:00", respectively.

[0188] For example, the health information estimation device 100 can use overview information (data structure list) that associates data structure with data arrangement patterns to determine the data structure corresponding to the data arrangement. For example, the health information estimation device 100 searches the data structure list, determines the data structure a associated with the data arrangement including data of categories such as water space data #1, #2 and external data #1, #2, and determines the determined data structure a as the data structure corresponding to "2024 / 3 / 1 8:00".

[0189] Furthermore, the health information estimation device 100 selects the health information estimation model shown in column 9 based on the data structure shown in column 8. For example, regarding the data corresponding to "2024 / 3 / 1 8:00", since the data structure is data structure a, the health information estimation device 100 selects model A as the health information estimation model corresponding to "2024 / 3 / 1 8:00". Similarly, regarding the data corresponding to "2024 / 3 / 1 12:00", "2024 / 3 / 2 9:30", and "2024 / 3 / 3 18:00", since the data structures are data structures b, c, and d, the health information estimation device 100 selects models B, C, and D respectively as the health information estimation models corresponding to "2024 / 3 / 1 12:00", "2024 / 3 / 2 9:30", and "2024 / 3 / 3 18:00". Furthermore, the health information estimation device 100 uses the selected model to estimate health information.

[0190] Additionally, for example, a health information estimation system 1 can also be used through... Figure 9 The processing is performed according to the process shown. Figure 9 This is a diagram illustrating an example of the processing flow in a health information estimation system. (Example:) Figure 9 As shown in the third processing flow PF3, in the health information estimation system 1, the health information estimation device 100 uses water space sensor data, which is then processed by the first data acquisition unit (e.g., with...). Figure 5 The toilet sensor data (labeled "toilet sensor data" in the figure) is acquired by the acquisition unit 131 (corresponding to the acquisition unit 131 in the figure). Additionally, the health information estimation device 100 acquires the toilet sensor data (labeled "toilet sensor data" in the figure) through the first data acquisition unit (e.g., with the acquisition unit 131 in the figure). Figure 5 (corresponding to the acquisition part 131 in the middle) to acquire external data.

[0191] Furthermore, the health information estimation device 100, based on the data arrangement of the acquired toilet sensor data and external data, determines the data through a data construction unit (e.g., with...). Figure 5 The decision unit 132 in the middle determines the data structure of the toilet sensor data and external data. Furthermore, the health information estimation device 100 uses a health information estimation unit (e.g., with...) Figure 5 The estimation unit 133 in the middle selects and determines the toilet sensor data and external data to construct a corresponding health information estimation model (model), and uses the selected model to estimate the user's health information.

[0192] Furthermore, in the above examples, such as datasets DS1 and DS2, the case where the data structure is determined solely based on whether or not data for each category is available, and the model is selected, is illustrated as an example. However, the health information estimation device 100 may also determine the data structure and select the model based on the amount and content of data for each category.

[0193] For example, Figure 10 As shown in the dataset DS3, the health information estimation device 100 determines the data structure of the toilet sensor data based on the data arrangement of the acquired toilet sensor data, and selects a health information estimation model (model) accordingly based on the determined data structure. Figure 10 This is a diagram illustrating an example of data processing. Figure 10 The dataset DS3 contains a group of data that associates five categories of data: the amount of odorless gas detected by the first detection sensor (odorless gas sensor), the amount of malodorous gas detected by the second detection sensor (malodorous gas sensor), and the amount, shape, and color of feces detected by the third detection sensor (feces sensor).

[0194] If the elements (squares) in the third row and thereafter of the columns corresponding to the amounts of odorless gases, malodorous gases, stool volume, stool shape, and stool color are not blank, then the corresponding data has been obtained. Regarding Figure 10 The data (records) in the third row indicate the amounts of odorless gas and malodorous gas obtained. That is, regarding... Figure 10 The data retrieved in the third row (records) indicates that no data was retrieved for the three categories of stool quantity, stool shape, and stool color. Additionally, regarding... Figure 10 The data (records) obtained in the third row show that data indicating the amount of odorless gas is 1 and the amount of malodorous gas is 1 was obtained.

[0195] about Figure 10 The data (records) in line 4 indicate the amounts of odorless gas and malodorous gas obtained. Additionally, regarding... Figure 10 The data (record) obtained in line 4 shows that data indicating an odorless gas quantity of 0.5 and an odorous gas quantity of 2 were obtained. Regarding... Figure 10 The data (records) obtained in line 5 indicate the amount of odorless gas, the amount of malodorous gas, the amount of stool, the shape of stool, and the color of stool. Additionally, regarding... Figure 10 The data (record) obtained in the third row shows that the amount of odorless gas is 1, the amount of malodorous gas is 1, the amount of feces is 4, the shape of feces is type 2, and the color of feces is brown.

[0196] exist Figure 10In this context, the health information estimation device 100 determines the data structure shown in the 6th column based on the data in columns 1 to 5 from the left in the dataset DS3. For example, the health information estimation device 100 regarding... Figure 10 The data (records) in the third row are arranged so that the amount of odorless gas is 1 and the amount of malodorous gas is 1. Therefore, Figure 10 The data structure for the data (record) in the third row is determined to be of gas type (data structure #11).

[0197] Additionally, for example, the health information estimation device 100 regarding Figure 10 The data (records) obtained in rows 4 and 5 will be constructed as either malodorous gas type (data construction #12) or gas + feces type (data construction #13).

[0198] Furthermore, the health information estimation device 100 selects the health information estimation model shown in the 7th column based on the data structure shown in the 6th column. For example, the health information estimation device 100 regarding... Figure 10 The data (record) retrieved in line 3, because the data structure is of gas type (data structure #11), is therefore used as... Figure 10 The gas type (model #11) is selected based on the corresponding health information estimation model obtained from the data (records) in the third row. Similarly, the health information estimation device 100... Figure 10 The data (records) obtained in rows 4 and 5 are constructed as malodorous gas type (data construction #12) and gas + feces type (data construction #13), therefore, they are used as... Figure 10 The health information estimation model corresponding to each acquired data (record) in rows 4 and 5 is selected based on the odor gas type (model #12) and gas + feces type (model #13). Furthermore, the health information estimation device 100 uses the selected model to estimate health information.

[0199] Furthermore, data from water space sensors and external sources is not limited to date and time information; it can be associated with various other types of information. For example, water space sensor data and external data can also be linked to the personal information of the user corresponding to that data. This will be discussed further later. Figure 11 This will be illustrated as an example.

[0200] For example, Figure 11 As shown in dataset DS4, the health information estimation device 100 uses data that associates the acquired water space sensor data and external data with the user's personal information (identification information such as user ID) corresponding to the data to determine the data structure, and selects a health information estimation model (model) accordingly based on the determined data structure. Figure 11 This is a diagram illustrating an example of data processing. Figure 11 The dataset DS4 contains a set of data that associates personal information and date / time information (i.e., time) with data obtained by the user on the date and time corresponding to that time, as indicated by the personal information. Therefore, the health information estimation device 100 can appropriately determine which user the estimated health information corresponds to. Furthermore, regarding other points, since it is related to... Figure 8 The content is the same as the descriptions above, so detailed explanations are omitted.

[0201] As described above, the data structure of the health information estimation system 1 is determined by the arrangement (combination) of data obtained at the date and time corresponding to the date and time information. For example, Figure 12 As shown in the dataset DS11, the health information estimation system 1 determines the data structure of the water space sensor data and external data based on the data arrangement of the water space sensor data and external data obtained at the date and time corresponding to the date and time information, and selects a health information estimation model (model) accordingly based on the determined data structure.

[0202] Figure 12 This is a diagram illustrating an example of data and processing. As mentioned above, in Figure 12 In this context, the arrangement (combination) of data corresponding to the horizontal data of dataset DS11, i.e., the combination of data categories, determines the data structure accordingly. For example, regarding the data corresponding to "2024 / 3 / 1 8:00" in the health information estimation device 100, since the data arrangement includes data of categories such as water space data #1, #2 and external data #1, #2, the data structure of the data corresponding to "2024 / 3 / 1 8:00" is determined to be data structure a.

[0203] Additionally, for example, the data corresponding to "2024 / 3 / 1 12:00" in the health information estimation device 100 is arranged as data structure b, since it includes data of categories such as water space data #2, #3 and external data #1, #2.

[0204] Furthermore, the above is merely an example; the health information estimation system 1 can also determine the data structure accordingly based on the date and time information corresponding to the date and time information and the data arrangement (combination) of the data obtained at that date and time. For example, Figure 13As shown in the dataset DS12, the health information estimation system 1 determines the data structure of the date and time, water space sensor data and external data based on the date and time corresponding to the date and time information and the data arrangement of the water space sensor data and external data obtained at that date and time, and selects the health information estimation model (model) accordingly based on the determined data structure.

[0205] Figure 13 This is a diagram illustrating an example of data and processing. As mentioned above, in Figure 13 In this context, the data arrangement (combination) of the date and time corresponding to the horizontal axis of dataset DS12, i.e., the combination of the date and time and the categories of the acquired data, determines the data structure accordingly. For example, regarding the data corresponding to "2024 / 3 / 1 8:00", the health information estimation device 100, since it includes data from categories such as 8:00 AM (early morning), water usage space data #1, #2, and external data #1, #2, determines the data structure of the data corresponding to "2024 / 3 / 1 8:00" as data structure a. 早晨 .

[0206] Additionally, for example, regarding the data corresponding to "2024 / 3 / 1 12:00" in the health information estimation device 100, since the data arrangement includes data categorized as 12:00 AM (noon), water usage space data #2, #3, and external data #1, #2, the data structure for the data corresponding to "2024 / 3 / 1 12:00" is determined to be data structure b. 中午 .

[0207] Furthermore, the health information estimation device 100 selects the health information estimation model shown in column 9 based on the data structure shown in column 8. For example, regarding the data corresponding to "2024 / 3 / 1 8:00", the health information estimation device 100 selects the health information estimation model shown in column 9 based on the data structure shown in column 8. 早晨 Therefore, Model A was chosen as the health information estimation model corresponding to "2024 / 3 / 1 8:00". 早晨 Similarly, the health information estimation device 100, regarding the data corresponding to "2024 / 3 / 1 12:00", uses data structure b. 中午 Therefore, Model B was chosen as the health information estimation model corresponding to "2024 / 3 / 1 12:00". 中午 Furthermore, the health information estimation device 100 uses the selected model to estimate health information.

[0208] In addition, the health information estimation system 1 can also supplement the data. For example, Figure 14As shown in the dataset DS13, the health information estimation system 1 supplements the data with past data, determines the data structure of the water space sensor data and external data based on the data arrangement of the supplemented data, and selects the health information estimation model (model) accordingly with the determined data structure.

[0209] Figure 14 This is a diagram illustrating an example of data and processing. In Figure 14 In this process, the health information estimation device 100 supplements the data with the latest data (corresponding to "2024 / 3 / 1 12:00") from the two sets of data corresponding to "2024 / 3 / 1 8:00" and the data corresponding to "2024 / 3 / 1 12:00". Specifically, the health information estimation device 100 supplements the data by using the water usage space data #1 obtained at "2024 / 3 / 1 8:00" as the data for the category of water usage space data #1 that was not obtained at the latest data ("2024 / 3 / 1 12:00").

[0210] In this case, the health information estimation device 100 determines the data structure corresponding to the supplemented "2024 / 3 / 1 12:00" as data structure e, since the data arrangement includes data of categories such as water usage space data #1 to #3 and external data #1 and #2. As described above, the health information estimation device 100 changes the data structure from the pre-supplement data structure b to the supplemented data structure e.

[0211] Furthermore, regarding the data corresponding to "2024 / 3 / 1 12:00", the health information estimation device 100 selects model E as the health information estimation model corresponding to "2024 / 3 / 1 12:00" because the data structure is data structure e. As described above, the health information estimation device 100 changes the model from the original model B to the revised model E. Moreover, the health information estimation device 100 uses the selected model to estimate health information.

[0212] Furthermore, in the above example, the health information estimation system 1 has been described as an example of selecting a model by determining the data structure accordingly based on the combination of data corresponding to the horizontal axis of the dataset DS, i.e., the combination of the categories of the acquired data, but it is not limited to this. For example, the health information estimation device 100 may also select a model by determining the data structure accordingly based on the data arrangement (combination) of the vertical axis of the dataset DS, i.e., the combination of the temporal sequence of the acquired data. Examples of this will be described below.

[0213] For example, Figure 15 As shown in the dataset DS14, the health information estimation device 100 determines the data structure of each category of data based on the data arrangement of the acquired data, and selects a health information estimation model (model) accordingly based on the determined data structure. Figure 15 This is a diagram illustrating an example of data processing. Figure 15 The DS14 dataset contains data sets that associate date and time information, i.e., time, with data obtained at the corresponding date and time. Figure 15 The diagram shows the acquisition of water space sensor data corresponding to categories #1 to #3.

[0214] exist Figure 15 Data for category #1 of water use space data was obtained at two times: the time corresponding to "2024 / 3 / 1 8:00" (hereinafter referred to as "the first acquisition time") and the time corresponding to "2024 / 3 / 3 18:00" (hereinafter referred to as "the fourth acquisition time"). That is, data for category #1 of water use space data was not obtained at the time corresponding to "2024 / 3 / 1 12:00" (hereinafter referred to as "the second acquisition time") and the time corresponding to "2024 / 3 / 2 9:30" (hereinafter referred to as "the third acquisition time").

[0215] The health information estimation device 100 performs time-series analysis based on information about the timing of data acquisition representing the category of water use spatial data #1, and determines the data structure of the data representing the category of water use spatial data #1. Figure 15 In the process, the health information estimation device 100 determines the data structure of the category of water use space data #1 as data structure a´ based on the timing analysis result of the acquisition of data at the first acquisition time and the fourth acquisition time.

[0216] Furthermore, regarding the category of water usage space data #2, the health information estimation device 100, based on the timing analysis result of the acquisition of data at the first and second acquisition times, determines the data structure of the category of water usage space data #2 as data structure b'. Additionally, regarding the category of water usage space data #3, the health information estimation device 100, based on the timing analysis result of the acquisition of data at the first to third acquisition times, determines the data structure of the category of water usage space data #3 as data structure c'.

[0217] For example, the health information estimation device 100 can also use overview information (data structure list) that associates data structure with the pattern of data arrangement for each data category to determine the data structure corresponding to the data arrangement. For example, the health information estimation device 100 retrieves the data structure list of water space data #1, determines the data structure a' associated with the data arrangement including the data of the first acquisition time and the fourth acquisition time, and determines the determined data structure a' as the data structure of the water space data #1.

[0218] Furthermore, the health information estimation device 100 selects a health information estimation model for each data category based on its respective data structure. For example, regarding the data for water space data #1, since the data structure is data structure a', the health information estimation device 100 selects model A' as the health information estimation model corresponding to water space data #1. Similarly, regarding the data corresponding to water space data #2 and #3, since the data structures are data structures b' and c', the health information estimation device 100 selects models B' and C' respectively as the health information estimation models corresponding to water space data #2 and #3. Moreover, the health information estimation device 100 uses the selected model to estimate health information. For example, if the data for the category of water space data #2 is foul-smelling gas data, the health information estimation device 100 can use the data for the category of water space data #2 obtained about the user and model B' to estimate the condition of the user's intestinal environment.

[0219] In addition, the situation regarding the acquisition of external data by Health Information Prediction System 1 is also related to... Figure 15 Similarly, the data arrangement (combination) corresponding to the longitudinal direction of the dataset DS, i.e., the temporal combination of the obtained data, can determine the data structure and select the model accordingly. This will be briefly explained below.

[0220] For example, Figure 16 As shown in the dataset DS15, the health information estimation device 100 determines the data structure of each category of data based on the data arrangement of the acquired data, and selects a health information estimation model (model) accordingly based on the determined data structure. Figure 16 This is a diagram illustrating an example of data processing. Figure 16 The DS15 dataset contains data sets that associate date and time information, i.e., time, with the data obtained at the corresponding date and time. Figure 16 The diagram shows the acquisition of water space sensor data of the categories corresponding to water space data #1 to #3, and external data of the categories #1 to #3.

[0221] In addition, regarding Figure 16 The water usage space data #1 to #3 in the data are related to... Figure 15 The water usage space data #1 to #3 are the same, therefore, the description is omitted. Figure 16 The data for category #3 of external data was acquired at two times: the time corresponding to "2024 / 3 / 2 9:30" (the third acquisition time) and the time corresponding to "2024 / 3 / 3 18:00" (the fourth acquisition time). That is, the data for category #3 of external data was not acquired at the time corresponding to "2024 / 3 / 1 8:00" (the first acquisition time) and the time corresponding to "2024 / 3 / 1 12:00" (hereinafter also referred to as the "second acquisition time").

[0222] The health information estimation device 100 performs timing analysis based on information indicating the timing of data acquisition for category #3 of external data, and determines the data structure of category #3 of external data. Figure 16 In the process, the health information estimation device 100 determines the data structure of the category of external data #3 as data structure d´ based on the timing analysis result of the acquisition of data at the 3rd acquisition time and the 4th acquisition time.

[0223] Furthermore, the health information estimation device 100 selects a health information estimation model for each data category based on its respective data structure. For example, regarding external data #3, since the data structure is data structure d´, the health information estimation device 100 selects model D´ as the health information estimation model corresponding to external data #3. The health information estimation device 100 then uses the selected model to estimate health information. Similarly, regarding external data #1 and #2, the health information estimation device 100 can determine the data structure based on the data arrangement of the data in that category, and select a model based on the determined data structure, just like with external data #3.

[0224] The health information estimation device 100 determines the data structure based on the combination of data corresponding to the horizontal relationship with the dataset DS, i.e., the combination of data categories obtained. When selecting a model, it can also use a model selected based on the most recent data for processing. Regarding this, using... Figure 17 Please provide an explanation. Figure 17 This is a diagram illustrating an example of data processing. Figure 17 The DS16 dataset in the dataset is due to its similarity to... Figure 8 The dataset is the same as DS2 in this dataset, so detailed explanations are omitted.

[0225] exist Figure 17In dataset DS16, the data obtained by the user at the time corresponding to "2024 / 3 / 3 18:00" is the most up-to-date data. Therefore, the health information estimation device 100 uses the health information estimation model, i.e., model D, to estimate the user's health information, and the data obtained at the time corresponding to "2024 / 3 / 3 18:00" is selected as the object of this health information estimation model.

[0226] Furthermore, the health information estimation device 100 determines the data structure based on the data arrangement (combination) corresponding to the longitudinal direction of the dataset DS, i.e., the temporal combination of the acquired data. When selecting a model, it can also use a model selected based on the category of each data point for processing. Regarding this, using... Figure 18 Please provide an explanation. Figure 18 This is a diagram illustrating an example of data processing. Figure 18 Since the dataset DS17 in this example is the same as the dataset DS14 in example 15, detailed descriptions are omitted.

[0227] exist Figure 18 In dataset DS17, model A' is selected for data on water use spatial data #1, model B' is selected for data on water use spatial data #2, and model C' is selected for data on water use spatial data #3. Therefore, the health information estimation device 100 uses model A' for data on water use spatial data #1, model B' for data on water use spatial data #2, and model C' for data on water use spatial data #3 to estimate health information.

[0228] Furthermore, the health information estimation device 100 determines the data structure based on the data arrangement (combination) corresponding to the longitudinal direction of the dataset DS, i.e., the temporal combination of the acquired data. When selecting a model, the model can be selected based on the overall category of the data. Regarding this, using... Figure 19 Please provide an explanation. Figure 19 This is a diagram illustrating an example of data processing. Figure 19 The DS21 dataset, apart from the health information inference model, is similar to... Figure 15 The dataset is the same as DS14 in the dataset, so detailed explanations other than the selection of the health information inference model are omitted.

[0229] exist Figure 19In the dataset DS21, the data structure for water space data #1 is determined as data structure a', the data structure for water space data #2 is determined as data structure b', and the data structure for water space data #3 is determined as data structure c'. Furthermore, the health information estimation device 100 selects model ABC' as the health information estimation model used in processing based on the combination of data structure a' for water space data #1, data structure b' for water space data #2, and data structure c' for water space data #3.

[0230] For example, the health information estimation device 100 can use a list of information (a health information estimation model list) that associates health information estimation models with combinations of data construction patterns to determine the health information estimation model corresponding to the combination pattern of data construction. For example, the health information estimation device 100 searches the health information estimation model list and determines the model ABC' associated with the combination of data construction a' of water use space data #1, data construction b' of water use space data #2, and data construction c' of water use space data #3, and selects the determined model ABC' as the health information estimation model to be used in the processing.

[0231] Furthermore, the health information estimation device 100 estimates health information using a selected model. For example, the health information estimation device 100 estimates health information using data from water space data #1, water space data #2, water space data #3, and model ABC'. For example, the health information estimation device 100 inputs data from water space data #1, water space data #2, and water space data #3 into model ABC', and estimates health information based on the information output by model ABC'.

[0232] Furthermore, the health information estimation device 100 can also weight (calculate weight values) the data for each category, reflecting the corrections implemented through these weight values, and use a model to make estimations. Regarding this, using... Figure 20 Please provide an explanation. Figure 20 This is a diagram illustrating an example of data processing. Figure 20 The DS22 dataset, besides adding weights related to each category, also has... Figure 19 The dataset DS21 is the same as the dataset in the dataset, so detailed explanations other than the weights are omitted.

[0233] exist Figure 20In dataset DS22, the data structure for water use spatial data #1 is determined as data structure a', the data structure for water use spatial data #2 is determined as data structure b', and the data structure for water use spatial data #3 is determined as data structure c'. Furthermore, the health information estimation device 100 calculates correction values, i.e., weight values, based on statistical values ​​of data from each category. For example, the more data the health information estimation device 100 obtains for each category, the larger the weight value is calculated. Figure 20 In the process, the health information estimation device 100 calculates the weight value of water use spatial data #1 as 0.3, the weight value of water use spatial data #2 as 0.3, and the weight value of water use spatial data #3 as 0.4.

[0234] Furthermore, the health information estimation device 100 selects model A as the health information estimation model used in the processing, based on a combination of data construction a' and weight value 0.3 for water use spatial data #1, data construction b' and weight value 0.3 for water use spatial data #2, and data construction c' and weight value 0.4 for water use spatial data #3. 0.3 B 0.3 C 0.4 Furthermore, the health information estimation device 100 uses the selected model to estimate health information. Additionally, the health information estimation device 100 can also select model ABC' as the health information estimation model used in processing, based on a combination of the data construction a' of water space data #1, the data construction b' of water space data #2, and the data construction c' of water space data #3, with a value of 0.4. In this case, the health information estimation device 100 can use the weight values ​​of water space data #1 (0.3), water space data #2 (0.3), and water space data #3 (0.4) as weight values ​​for model ABC'. For example, the health information estimation device 100 can input the data of water space data #1, water space data #2, and water space data #3, along with the weight values ​​of water space data #1 (0.3), water space data #2 (0.3), and water space data #3 (0.4), into model ABC', and estimate health information based on the information output by model ABC'.

[0235] Furthermore, the health information estimation device 100 determines the data structure based on the data arrangement (combination) corresponding to the longitudinal direction of the dataset DS, i.e., the temporal combination of the acquired data. When selecting a model, the model can be selected based on the combination of the categories of water space sensor data and the categories of external data. Regarding this, using... Figure 21 Please provide an explanation. Figure 21 This is a diagram illustrating an example of data processing. Figure 21The DS23 dataset, apart from its health information inference model, is similar to... Figure 16 The dataset is the same as DS15 in this dataset, so detailed explanations other than the choice of the health information inference model are omitted.

[0236] exist Figure 21 In dataset DS23, the data structure of water space data #1 is determined as data structure a', and the data structure of external data #3 is determined as data structure d'. Furthermore, the health information estimation device 100 selects model AD as the health information estimation model used in processing based on the combination of data structure a' of water space data #1 and data structure d' of external data #3. The health information estimation device 100 then uses the selected model to estimate health information.

[0237] Additionally, for example, a health information estimation system 1 can also be used through... Figure 22 The processing flow shown is executed. Figure 22 This is a diagram illustrating an example of the processing flow in a health information estimation system. Furthermore, Figure 22 In the fourth processing step PF11, besides obtaining supplementary data from the user, it also... Figure 9 The third processing flow PF3 is the same as that in the previous process, so detailed explanations of parts other than obtaining supplementary data from the user are omitted.

[0238] When determining the data structure, the health information estimation device 100 requests the user to supplement the data if necessary. Furthermore, the health information estimation device 100 processes the supplementary data obtained from the user by appending it to external data sources.

[0239] As described above, the health information estimation system 1 is not limited to fecal matter; it can estimate a user's health information based on data from all water-using space sensors. For example, the health information estimation system 1 can estimate a user's health information based on various categories of data, such as fecal and urinary data obtainable in the toilet, and daily activity data obtainable in the bathroom (bathtub) and kitchen. In the above processing, the health information estimation system 1 structures the various data and processes them through statistics or algorithms in order to estimate health information.

[0240] For example, in existing systems, when it is determined that additional data beyond the existing excrement data is needed to presume health information, the system can request additional data from the user via a smartphone or other device. However, if the user does not respond to the request, the health information cannot be determined, thus leaving room for improvement. Therefore, in the health information presumption system 1, when it is determined that additional data beyond the already obtained data is needed to presume health information, the system requests additional data from the user, thereby allowing for appropriate supplementation and processing of the data.

[0241] <4-1. Examples of other information inferences>

[0242] The above information estimation is merely an example; the health information estimation system 1 is not limited to the above information and can also estimate various types of information. Several examples are provided below. The following is an example of an information estimation system where the health information estimation system 1 estimates at least one of the user's health information and the service information provided to the user based on data detected by sensors deployed in the water-using space.

[0243] In the example below, the health information estimation system 1 determines the data structure based on a data arrangement or data group formed by associating data detected by sensors deployed in the water-using space according to date and time information. Furthermore, when estimating the user's health information or service information provided to the user based on the data, the health information estimation system 1 selects and uses a model corresponding to the data structure from multiple information estimation models.

[0244] First, based on Figure 23 The example shown illustrates an example of information processing performed by the information inference system, namely the health information inference system 1. Figure 23 This is a diagram illustrating an example of data and processing. Furthermore, explanations of points identical to those described above are appropriately omitted.

[0245] For example, Figure 23 As shown in the dataset DS31, the health information estimation device 100 determines the data structure of the acquired water space sensor data based on the data arrangement of the data, and selects an estimation model accordingly based on the determined data structure. Furthermore, the estimation model mentioned here may be the same as or different from the health information estimation model described above.

[0246] For example, an inference model can be a model that takes as input data constructed from data associated with the inference model and outputs at least one of the following: user health information inferred from the data, or service information provided to the user. In this case, the inference model can be a machine learning model (AI model) trained using machine learning.

[0247] For example, an inference model can be an AI model that, given data about a user with a data structure associated with it, outputs at least one of the inferred health information or service information about that user based on the input of that data. For example, an inference model can be an AI model that, based on input data with a data structure associated with it, outputs a score (value) representing a certain health state (e.g., gut health, nocturnal urination, overactive bladder, menstrual cycle, stress, weakness, excessive salt intake, etc.).

[0248] Alternatively, an inference model can also be an AI model that outputs service information to a user who is the source of that data, based on input data constructed from data associated with the inference model. Furthermore, an inference model can be any AI model as long as it can output at least one of the inferred health information or service information about the user. For example, an inference model could be an LLM (Large Language Model). As mentioned above, an inference model can be any AI model, such as a generative AI like an LLM, as long as it can output the desired information.

[0249] Alternatively, an inference model can also be an AI model that outputs information representing a classification of the user's health status corresponding to the input, based on input data constructed from data associated with the inference model. For example, an inference model can be a classifier (identifier) ​​that outputs information such as whether the user's health status corresponds to (is classified as) good (health status), intestinal status, nocturnal urination, overactive bladder, menstrual cycle, stress, weakness, excessive salt intake, etc., based on input data constructed from data associated with the inference model. In this case, excessive salt intake could also mean excessive salt consumption. As mentioned above, an inference model can be a function that generates new information based on data, such as an AI model. The term "function" here, as long as it generates new information based on data, is not limited to AI models but includes various things such as programs and algorithms.

[0250] Furthermore, the presumption model is not limited to the functions mentioned above. Any model can be used as long as it can presume at least one of the user's health information or the service information provided to the user. For example, the presumption model could be information indicating the health status of a user who has obtained data with a data structure associated with the presumption model. In this case, the presumption model could be information (type information) indicating the type of health status a user who has obtained data with a data structure associated with the presumption model is classified into. For example, the presumption model could be type information indicating whether a user who has obtained data with a data structure associated with the presumption model is classified as having good health, good bowel function, frequent urination at night, overactive bladder, menstrual cycle, stress, weakness, excessive salt intake, etc.

[0251] exist Figure 23 In this system, the health information estimation system 1 processes data (urine data) representing the composition of urine detected by a sensor (also called a "urine sensor") that detects the composition of urine in excrement discharged into the toilet device 50. In this case, the sensor device 5 has a urine sensor that detects the composition of urine. For example, the urine sensor can be installed on the toilet seat of the toilet device 50. Furthermore, the urine sensor is a known sensor, and detailed description is omitted; however, any sensor can be used as long as it can detect the desired composition of urine.

[0252] In addition, Figure 23 The example shown illustrates the use of urine data as biological data, but the health information estimation system 1 is not limited to using urine data; it can also use various biological data for the processing described below. For example, the health information estimation system 1 is not limited to urine; it can also use various data (biological data) such as feces (defecation), fecal gas, skin gas, and blood flow. For example, the health information estimation system 1 can process multiple types of biological data, including feces (defecation), urine, fecal gas, skin gas, and blood flow. The sensor device 5 has various sensors for acquiring the biological data used by the health information estimation system 1 in the processing.

[0253] Furthermore, the various sensors included in the sensor device 5 can be configured in any location, as long as they are capable of performing the desired detection. For example, the various sensors included in the sensor device 5 are not limited to the toilet device 50, but can also be configured in, for example, a remote control that operates the sensor device 5. For example, a sensor that detects biological data can be configured in the remote control, detecting the user's biological data based on the user's finger touching the remote control. In addition, the various sensors included in the sensor device 5 can be either contact sensors or non-contact sensors.

[0254] exist Figure 23 In the example, as shown in the column labeled as the first detection sensor, data on urine-related components such as urea, sodium, and potassium (also known as "urine component data") are obtained based on biological data detected by the urine sensor. Figure 23 The dataset DS31 contains a group of data that associates date and time information, i.e., time, with urine component data obtained at the corresponding date and time.

[0255] exist Figure 23 In the health information estimation system 1, the health information estimation device 100 determines the data structure shown in the 5th column based on the arrangement of the data (records) from the leftmost columns 1 to 4 in the dataset DS31, namely the time and urine component data. For example, the health information estimation device 100 determines the data structure based on the time of each data arrangement in the dataset DS31.

[0256] For example, regarding the data corresponding to "2024 / 11 / 19 8:00", the health information estimation device 100 determines the data structure of the data corresponding to "2024 / 11 / 19 8:00" as data structure #31 because the data was obtained at 8:00 AM, i.e., in the morning. For example, data structure #31 is a urine-morning type data structure.

[0257] Additionally, for example, regarding the data corresponding to "2024 / 11 / 19 12:00", the health information estimation device 100 determines the data structure of the data corresponding to "2024 / 11 / 19 12:00" as data structure #32 because the data was obtained at 12:00 (12:00 AM), which is noon. For example, data structure #32 is a urine-noon type data structure.

[0258] For example, regarding the data corresponding to "2024 / 11 / 20 8:00", the health information estimation device 100 determines the data structure for the data corresponding to "2024 / 11 / 20 8:00" as data structure #31, since the data was obtained at 8:00 AM, i.e., in the morning. For example, data structure #31 is a urine-morning type data structure. Furthermore, the above is just one example; the health information estimation device 100 is not limited to time and can also determine the data structure based on urine component data.

[0259] Furthermore, the health information estimation device 100 selects the estimation model shown in column 6 based on the data structure shown in column 5. For example, regarding the data corresponding to "2024 / 11 / 19 8:00", the health information estimation device 100 selects model #31 as the estimation model corresponding to "2024 / 11 / 19 8:00" since the data structure is data structure #31. For example, model #31 is the estimation model used regarding urine-morning type (data structure #31).

[0260] Furthermore, regarding the data corresponding to "2024 / 11 / 19 12:00", the health information estimation device 100 selects model #31 as the estimation model corresponding to "2024 / 11 / 19 12:00" because the data structure is data structure #32. For example, model #32 is the estimation model used regarding urine-midday type (data structure #32). Additionally, regarding the data corresponding to "2024 / 11 / 20 8:00", the health information estimation device 100 selects model #31 as the estimation model corresponding to "2024 / 11 / 20 8:00" because the data structure is data structure #31.

[0261] For example, the health information estimation device 100 uses model #31 to estimate the user's health information or the service information provided to the user based on data corresponding to "2024 / 11 / 19 8:00". For instance, if model #31 is an estimation model that outputs information indicating whether the user is classified as having excessive salt intake, the health information estimation device 100 inputs the data corresponding to "2024 / 11 / 19 8:00" into model #31. Furthermore, model #31, upon receiving the data corresponding to "2024 / 11 / 19 8:00", outputs information indicating that the user of the data corresponding to "2024 / 11 / 19 8:00" is classified as having excessive salt intake. In this case, the health information estimation device 100 estimates that the user of the data corresponding to "2024 / 11 / 19 8:00" has excessive salt intake.

[0262] For example, if model #31 is an inferred model that outputs service information provided to the user, the health information inference device 100 inputs data corresponding to "2024 / 11 / 19 8:00" into model #31. Moreover, model #31, which has been input with data corresponding to "2024 / 11 / 19 8:00", outputs service information that indicates health-related suggestions to the user, such as "Due to excessive salt content, please use low-sodium soy sauce".

[0263] Furthermore, model #31 is not limited to the above-mentioned output content and can also output various information. For example, model #31 can output URLs (Uniform Resource Locators) of EC (Electronic Commerce) sites selling products such as soy sauce. Thus, the health information estimation device 100 estimates the service information that should be provided to users of the data corresponding to "2024 / 11 / 19 8:00". As described above, the health information estimation device 100 performs estimation processing using a model selected based on data construction about the data obtained from the user. Therefore, the health information estimation system can appropriately estimate information related to a person's health.

[0264] Furthermore, the health information estimation device 100 can also send information based on estimation processing to the user. For example, the health information estimation device 100 can send at least one of the user's health information estimated through estimation processing or service information provided to the user to the terminal device 10 used by the user. For example, the health information estimation device 100 can send service information such as "Please use low-sodium soy sauce due to excessive salt intake" estimated by user U1 through estimation processing to the terminal device 10 used by user U1. In addition, for example, the health information estimation device 100 can send service information including the URL of an e-commerce site selling products such as soy sauce to the terminal device 10 used by user U1.

[0265] exist Figure 23 The example shown illustrates inference processing based on the arrangement of individual data points, but the health information inference system 1 can also process data based on multiple records. Regarding this, [the system uses...] Figure 24 Please provide an explanation. Figure 24 This is a diagram illustrating an example of data and processing. Furthermore, in Figure 23 Points identical to those above will be omitted as appropriate. Hereinafter, multiple data records may be referred to as a data group. Figure 24 In the dataset DS32, the column labeled Presumed Model 2 represents the presumed model that uses multiple data (records) as objects.

[0266] exist Figure 24 In this context, the health information estimation device 100 selects model #30 as the estimation model for the data set corresponding to "2024 / 11 / 19 8:00" to "2024 / 11 / 23 8:00". For example, model #30 is a one-week model (estimated model) used for multiple data sets (data sets) obtained within a multi-day period (within one week).

[0267] For example, the health information estimation device 100 uses model #30 to estimate the user's health information or the service information provided to the user based on the data set corresponding to "2024 / 11 / 19 8:00" to "2024 / 11 / 23 8:00". For example, if model #30 is an estimation model that outputs health information, the health information estimation device 100 inputs the data set corresponding to "2024 / 11 / 19 8:00" to "2024 / 11 / 23 8:00" into model #30. Moreover, model #30, which is input with the data set corresponding to "2024 / 11 / 19 8:00" to "2024 / 11 / 23 8:00", outputs health information representing the health status of the user who is the source of the data set.

[0268] Therefore, the health information estimation device 100 estimates the health information of users who are the source of the data set corresponding to "2024 / 11 / 19 8:00" to "2024 / 11 / 23 8:00". Additionally, the health information estimation device 100 can also estimate service information to be provided to users using model #30 regarding the data set corresponding to "2024 / 11 / 19 8:00" to "2024 / 11 / 23 8:00". Furthermore, regarding the processing related to the estimation of information and the provision of the estimated information, since... Figure 23 The details are the same, so a detailed explanation is omitted.

[0269] Furthermore, the health information estimation system 1 is not limited to the above processing; for example, it can also perform... Figure 25 The processing is as shown. Figure 25 This is a diagram illustrating an example of data and processing. In Figure 25 The text illustrates, as an example, the use of urine data obtained initially (first time) of the day as the data used in the processing. Furthermore, in... Figure 23 and Figure 24 Points that are the same as those mentioned above have been appropriately omitted from the explanation.

[0270] exist Figure 25 In the example, as shown in the column labeled "Detection Sensor 1," data on urinary sodium and potassium composition obtained based on biological data detected by a urine sensor is illustrated. However, this can also be combined with... Figure 23 and Figure 24 Similarly, urea is included in the urine composition data. Figure 25 The dataset DS33 contains a group of data that associates date and time information, i.e., time, with urine component data obtained at the corresponding date and time.

[0271] exist Figure 25In this process, the health information estimation device 100 determines the data structure based on the time of each data arrangement in the dataset DS33. For example, regarding the data corresponding to "2024 / 11 / 19 8:00", since the data arrangement is based on the data initially acquired at 8:00 AM (November 19th), the health information estimation device 100 determines the data structure corresponding to "2024 / 11 / 19 8:00" as data structure #33. For example, data structure #33 is the data structure adopted as the data used in processing (also called "adopted data structure").

[0272] Additionally, for example, regarding the data corresponding to "2024 / 11 / 19 12:00", the health information estimation device 100 determines the data structure for the data corresponding to "2024 / 11 / 19 12:00" as data structure #34 because it is the data arrangement of the second data obtained at 12:00 (12:00 AM) on November 19th. For example, data structure #34 is a data structure that is not used as data in the processing (also known as "not using data structure").

[0273] For example, regarding the data corresponding to "2024 / 11 / 20 8:00", since the data arrangement is based on the data initially acquired at 8:00 AM (November 20th), the data structure for the data corresponding to "2024 / 11 / 20 8:00" is determined to be data structure #33. Similarly, regarding the data corresponding to "2024 / 11 / 21 9:00", since the data arrangement is based on the data initially acquired at 9:00 AM (November 21st), the data structure for the data corresponding to "2024 / 11 / 21 9:00" is determined to be data structure #33.

[0274] Furthermore, for example, regarding the data corresponding to "2024 / 11 / 22 12:00", the health information estimation device 100 determines the data structure of the data corresponding to "2024 / 11 / 22 12:00" as data structure #33 because it is the data arrangement of the data initially obtained at 12:00 (12:00 AM) on November 22nd. As described above, in Figure 25 In the example, even data acquired at the same time can sometimes have different data structures depending on the circumstances under which the data was acquired on that day. For instance, the health information estimation device 100 determines the data structure corresponding to "2024 / 11 / 23 8:00" as data structure #33 because it is based on the data arrangement of the data initially acquired at 8:00 AM on November 23rd.

[0275] exist Figure 25 In the dataset DS33, the column labeled as the presumptive model indicates the following presumptive model, which uses the data associated with data construction #33 (using data construction) as the data used in the processing as an object.

[0276] exist Figure 25 In this context, the health information estimation device 100 selects model #35 as the estimation model for the data set (also referred to as the "data set using data construction") associated with data construction #33 (using data construction) (lines 3, 5-8). For example, model #35 is an estimation model (sodium-potassium model) used for the data set using data construction. For example, model #35 is an estimation model that estimates information based on sodium and potassium in urine.

[0277] For example, Model #35 could be a model that infers a user's health information, such as blood pressure, based on the sodium-to-potassium ratio (also known as the "Na / K ratio") in a dataset constructed from data. For example, Model #35 could be a model that infers higher blood pressure in a user if the Na / K ratio in the constructed dataset is above a specified threshold. Alternatively, Model #35 could be a model that infers hypertension in a user if the Na / K ratio in the constructed dataset is above a specified threshold.

[0278] For example, the health information estimation device 100 regarding Figure 25 The data set constructed using data is used to infer the user's health information or provide the user with service information using model #35. For example, if model #35 is an inference model that outputs health information, the health information inference device 100 will... Figure 25 The data set constructed using the data is input into model #35. Furthermore, the input... Figure 25 Model #35, which uses data to construct a data set, outputs health information representing the health status of the user who is the source of the data set. Thus, the health information estimation device 100 outputs health information representing the health status of the user who is the source of the data set. Figure 25 The data set is constructed using data to estimate the health information of the user from the source.

[0279] For example, regarding urine, the information obtained varies depending on factors such as urine composition, data acquisition time, and data acquisition period (once, once a day, once a week), and preferably, the optimal estimation model also differs accordingly. Therefore, as described above, the health information estimation device 100 can provide at least one of health information and service information based on various urine-related information and using the optimal estimation model corresponding to the data used in processing.

[0280] As described above, the data structure described in this application can include various structures. For example, a data structure can be based on a pattern of the presence or absence of data for each item (category) in the data arrangement. Alternatively, a data structure can be based on a pattern of date and time information (moments) in the data arrangement. Furthermore, a data structure can be based on a pattern of the values ​​(numerical values) of each item (category) in the data arrangement. Additionally, a data structure can be based on a pattern of the ratios of the numerical values ​​(components, etc.) of each item (category) in the data arrangement. Furthermore, a data structure can be based on a pattern combining the above classifications. As described above, a data structure can also be a structure representing a classification formed by patterns based on various information.

[0281] <5. Others>

[0282] As mentioned above, it is not limited to a single user (in Figure 1 In the case of user U1 using the water-using space PS, the health information estimation system 1 can also be applied to situations where multiple users use the same water-using space PS. That is, the health information estimation system 1 is not limited to water-using spaces PS configured in a residence occupied by one person, but can also be applied to users of water-using spaces PS configured in facilities (buildings, etc.) shared by an uncertain number of people. An example of this situation will be briefly explained.

[0283] As described above, assuming multiple users utilize the water-using space PS, the health information estimation system 1 can determine the users of the water-using space PS in any manner. In this case, Figure 1 In this system, the user can be determined using an operating device installed in the water-using space (PS). For example, the health information estimation system 1 determines the user by having the user operate the operating device of the water-using space (PS) to designate the user. Figure 1 In the example, user U1 operates the operating device of toilet device 50, and by designating user U1, health information estimation system 1 determines that the user is user U1.

[0284] For example, the health information estimation system 1 determines the user by having the user specify a user from the user group displayed on the operating device of the water-using space PS. Additionally, in Figure 1 In this process, user U1's terminal device 10 pairs with the operating device of the water-using space PS, for example, via Bluetooth, thereby confirming that user U1 is a user of the water-using space PS. Alternatively, the user can be determined by the detection results of various sensors installed in the water-using space PS. Furthermore, the above is just one example; as long as the user of the water-using space PS can be determined, the user can be determined through any processing.

[0285] As described above, when the user of the water-using space PS can be identified, each water-using space PS, such as toilet space PS1, can be a space located in any place. For example, toilet space PS1 can be a so-called public toilet. For example, toilet space PS1 is not limited to a residence where family members live, but can also be a toilet space in facilities where many people live (nursing homes, senior apartments, etc.), office buildings, shops such as shopping malls (department stores), amusement parks, stadiums, parks, parking lots, etc. As described above, water-using space PS can be a water-using space located in any place as long as the processing performed by the health information estimation system 1 can be applied.

[0286] Further effects and variations can be readily derived by those skilled in the art. Therefore, the invention is not limited to the specific details and representative embodiments described above. Thus, various modifications can be made without departing from the spirit or scope of the comprehensive concept of the invention as defined by the appended claims and their equivalents.

[0287] The above-described embodiments and variations may have the structures shown below, but are not limited to the following structures. (1)

[0289] A health information estimation system, characterized in that it has: The data acquisition unit acquires data detected by sensors configured in the water-using space; A data construction determination unit determines the data construction based on a data arrangement formed by associating data obtained through the data acquisition unit according to date and time information; and A health information estimation unit, which estimates the user's health information based on the data arrangement. The health information estimation unit selects a model corresponding to the data construction from multiple health information estimation models to estimate the user's health status. (2)

[0291] A health information estimation system, characterized in that it has: The first data acquisition unit acquires data detected by sensors configured in the water-using space; The second data acquisition unit acquires external data; A data structure determination unit determines the data structure based on a data arrangement formed by associating at least one of the data acquired by the first data acquisition unit and the second data acquisition unit according to date and time information; and A health information estimation unit, which estimates the user's health information based on the data arrangement. The health information estimation unit selects a model corresponding to the data construction from multiple health information estimation models to estimate the user's health status. (3)

[0293] A health information estimation system, characterized in that it has: The first data acquisition unit acquires data detected by sensors configured in the toilet space; The second data acquisition unit acquires external data; A data structure determination unit determines the data structure based on a data arrangement formed by associating at least one of the data acquired by the first data acquisition unit and the second data acquisition unit according to date and time information; and A health information estimation unit, which estimates the user's health information based on the data arrangement. The health information estimation unit selects a model corresponding to the data construction from multiple health information estimation models to estimate the user's health status. (4)

[0295] According to the health information estimation system described in (1), the system is characterized in that, The data construction decision unit determines the data construction based on the data arrangement formed by associating the data obtained by the data acquisition unit with personal information or date information. (5)

[0297] The health information estimation system described in any one of (1) to (4) is characterized in that, The data construction determination unit determines the data construction based on a combination of at least one data element of the data arrangement. (6)

[0299] The health information estimation system described in any one of (1) to (5) is characterized in that, The data construction decision unit uses past data arrangements within a specified date and time range to supplement the missing data arrangements in the latest data arrangement, and determines the data construction based on the supplemented data arrangements. (7)

[0301] The health information estimation system described in (1) or (4) is characterized in that, The data construction decision unit determines the data construction based on the time-series analysis results of each data obtained by the data acquisition unit within a specified date and time range. (8)

[0303] The health information estimation system described in (2) or (3) is characterized in that, The data construction decision unit determines the data construction based on the time sequence analysis results of each data obtained by the first data acquisition unit and the second data acquisition unit within a specified date and time range. (9)

[0305] The health information estimation system described in any one of (1) to (8) is characterized in that, The health information estimation unit selects the model corresponding to the latest data construction from the plurality of health information estimation models to estimate the user's health status. (10)

[0307] The health information estimation system described in any one of (1), (4), and (7) is characterized in that, The health information estimation unit, in accordance with the data construction based on the time-series analysis results of each data obtained by the data acquisition unit within a specified date and time range, selects a model from the plurality of health information estimation models to estimate the user's health status. (11)

[0309] The health information estimation system described in any one of (1), (4), (7), and (10) is characterized in that, The health information estimation unit, in conjunction with multiple data structures based on the time-series analysis results of each data obtained by the data acquisition unit within a specified date and time range, selects a model from the multiple health information estimation models to estimate the user's health status. (12)

[0311] The health information estimation system described in any one of (1), (4), (7), (10), and (11) is characterized in that, The health information estimation unit corrects the health information estimation model based on the correction value calculated corresponding to the data obtained by the data acquisition unit. (13)

[0313] The health information estimation system described in any one of (2), (3), and (8) is characterized in that, The health information estimation unit, in accordance with multiple data structures based on the time-series analysis results of each data obtained by the first data acquisition unit and the second data acquisition unit within a specified date and time range, selects a model from the multiple health information estimation models to estimate the user's health status. (14)

[0315] The health information estimation system described in any one of (1) to (13) is characterized in that, The health information estimation unit estimates the user's health information at a predetermined period. (15)

[0317] The health information estimation system described in any one of (2), (3), (8), and (13) is characterized in that, In order to fill in the gaps in the data arrangement within the specified date and time range, the data construction decision unit prompts the user to supplement the external data. (16)

[0319] An information estimation system, characterized in that it has: The data acquisition unit acquires data detected by sensors configured in the water-using space; A data construction decision unit determines the data construction based on a data arrangement or data group formed by associating data obtained by the data acquisition unit according to date and time information; and The information estimation unit estimates the user's health information or the service information provided to the user based on the data obtained by the data acquisition unit. The information estimation unit selects a model from multiple information estimation models that corresponds to the data construction. (17)

[0321] The system for estimating information as described in (16) is characterized in that, The sensor is used to detect the composition of urine in excrement discharged into the toilet system. The data acquisition unit acquires urine component data detected by the sensor. The data construction decision unit determines the urine data construction based on the urine component arrangement formed by associating the urine component data according to date and time information. (18)

[0323] The system for estimating information as described in (16) is characterized in that, The sensor is installed in the toilet space to acquire the user's biological data. The data acquisition unit acquires biological data detected by the sensor. The data construction decision unit determines the biological data construction based on the biological data arrangement formed by associating the biological data according to date and time information.

[0324] Explanation of the label

[0325] 1. Health Information Prediction System (Information Prediction System)

[0326] 2 Server Devices

[0327] 5. Sensor Device

[0328] 10. Terminal devices (equipment)

[0329] 50 Toilet installations

[0330] 100 Health Information Estimation Device

[0331] 110 Ministry of Communications

[0332] 120 Storage Department

[0333] 121 Data Storage Department

[0334] 122 Model Information Storage Department

[0335] 123 User Information Storage Department

[0336] 130 Control Department

[0337] 131 Acquisition Department

[0338] 132 Decision Department

[0339] 133 Presumption Department

[0340] 134 Sending Department

Claims

1. A health information estimation system, characterized in that, have: The data acquisition unit acquires data detected by sensors configured in the water-using space; The data construction decision unit determines the data construction based on the data arrangement formed by associating the data obtained by the data acquisition unit according to date and time information; as well as A health information estimation unit, which estimates the user's health information based on the data arrangement. The health information estimation unit selects a model corresponding to the data construction from multiple health information estimation models to estimate the user's health status.

2. A health information estimation system, characterized in that, have: The first data acquisition unit acquires data detected by sensors configured in the water-using space; The second data acquisition unit acquires external data; The data construction decision unit determines the data construction based on the data arrangement formed by associating at least one of the data obtained by the first data acquisition unit and the second data acquisition unit according to date and time information; as well as A health information estimation unit, which estimates the user's health information based on the data arrangement. The health information estimation unit selects a model corresponding to the data construction from multiple health information estimation models to estimate the user's health status.

3. A health information estimation system, characterized in that, have: The first data acquisition unit acquires data detected by sensors configured in the toilet space; The second data acquisition unit acquires external data; The data construction decision unit determines the data construction based on the data arrangement formed by associating at least one of the data obtained by the first data acquisition unit and the second data acquisition unit according to date and time information; as well as A health information estimation unit, which estimates the user's health information based on the data arrangement. The health information estimation unit selects a model corresponding to the data construction from multiple health information estimation models to estimate the user's health status.

4. The health information estimation system according to claim 1, characterized in that, The data construction decision unit determines the data construction based on the data arrangement formed by associating the data obtained by the data acquisition unit with personal information or date information.

5. The health information estimation system according to claim 1, characterized in that, The data construction determination unit determines the data construction based on a combination of at least one data element of the data arrangement.

6. The health information estimation system according to claim 1, characterized in that, The data construction decision unit uses past data arrangements within a specified date and time range to supplement the missing data arrangements in the latest data arrangement, and determines the data construction based on the supplemented data arrangements.

7. The health information estimation system according to claim 1, characterized in that, The data construction decision unit determines the data construction based on the time-series analysis results of each data obtained by the data acquisition unit within a specified date and time range.

8. The health information estimation system according to claim 2, characterized in that, The data construction decision unit determines the data construction based on the time sequence analysis results of each data obtained by the first data acquisition unit and the second data acquisition unit within a specified date and time range.

9. The health information estimation system according to claim 1, characterized in that, The health information estimation unit selects the model corresponding to the latest data construction from the plurality of health information estimation models to estimate the user's health status.

10. The health information estimation system according to claim 1, characterized in that, The health information estimation unit, in accordance with the data construction based on the time-series analysis results of each data obtained by the data acquisition unit within a specified date and time range, selects a model from the plurality of health information estimation models to estimate the user's health status.

11. The health information estimation system according to claim 1, characterized in that, The health information estimation unit, in conjunction with multiple data structures based on the time-series analysis results of each data obtained by the data acquisition unit within a specified date and time range, selects a model from the multiple health information estimation models to estimate the user's health status.

12. The health information estimation system according to claim 1, characterized in that, The health information estimation unit corrects the health information estimation model based on the correction value calculated corresponding to the data obtained by the data acquisition unit.

13. The health information estimation system according to claim 2, characterized in that, The health information estimation unit, in accordance with multiple data structures based on the time-series analysis results of each data obtained by the first data acquisition unit and the second data acquisition unit within a specified date and time range, selects a model from the multiple health information estimation models to estimate the user's health status.

14. The health information estimation system according to claim 1, characterized in that, The health information estimation unit estimates the user's health information at a predetermined period.

15. The health information estimation system according to claim 2, characterized in that, In order to fill in the gaps in the data arrangement within the specified date and time range, the data construction decision unit prompts the user to supplement the external data.

16. An information estimation system, characterized in that, have: The data acquisition unit acquires data detected by sensors configured in the water-using space; The data construction decision unit determines the data construction based on the data arrangement or data group formed by associating the data obtained by the data acquisition unit according to date and time information. as well as The information estimation unit estimates the user's health information or the service information provided to the user based on the data obtained by the data acquisition unit. The information estimation unit selects a model from multiple information estimation models that corresponds to the data construction.

17. The information estimation system according to claim 16, characterized in that, The sensor is used to detect the composition of urine in excrement discharged into the toilet system. The data acquisition unit acquires urine component data detected by the sensor. The data construction decision unit determines the urine data construction based on the urine component arrangement formed by associating the urine component data according to date and time information.

18. The information estimation system according to claim 16, characterized in that, The sensor is installed in the toilet space to acquire the user's biological data. The data acquisition unit acquires biological data detected by the sensor. The data construction decision unit determines the biological data construction based on the biological data arrangement formed by associating the biological data according to date and time information.

Images