Information processing system and method

The system addresses the limitation of existing pet health management by using behavioral data analysis to improve health risk detection and solution implementation in pet health management systems.

JP7594305B2Active Publication Date: 2024-12-04RABO INC
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
JP2022522194
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Priority Date
2020-05-25
Filing Date
2021-05-13
Publication Date
2024-12-04
Estimated Expiration
2041-05-13

AI Technical Summary

Technical Problem

Existing pet health management systems primarily focus on excretion and feeding, neglecting the importance of behavioral data in understanding a pet's health condition.

Method used

An information processing system that utilizes sensors to acquire measurement data, applies machine learning to detect changes in pet behavior, and provides health risk assessments and solution proposals based on this data.

Benefits of technology

Enhances the accuracy of pet health condition understanding by analyzing behavioral data, allowing for timely health risk detection and effective solution implementation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007594305000001
    Figure 0007594305000001
  • Figure 0007594305000002
    Figure 0007594305000002
  • Figure 0007594305000003
    Figure 0007594305000003
Patent Text Reader

Abstract

[Problem] To detect changes in the state of a pet's health. [Solution] This information processing system comprises: a measured data-acquiring unit 21 for acquiring measured data relating to a pet as a target of detection; a risk assessment unit 23 (detection unit 25) for detecting changes in the state of the pet on the basis of the acquired measured data; and a notification unit 24 (output unit 27) for outputting information corresponding to the detected change in the pet's state.
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present disclosure relates to an information processing system and method for providing a service for understanding the health condition of a pet. [Background technology]

[0002] Patent Document 1 describes the use of an animal litter box capable of calculating the weight of pet waste to determine abnormalities in the health of pets. Patent Document 2 describes a management server that determines the composition of ingredients contained in food provided to pets based on the health condition obtained from a sensor attached to the pet and outputs the information on the composition of ingredients. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 6585523 [Patent Document 2] JP 2020-5534 A Summary of the Invention [Problem to be solved by the invention]

[0004] The above patent documents do not mention health management based on animal behavior other than excretion and feeding.

[0005] Therefore, an object of the present disclosure is to provide an information processing system and method capable of detecting changes in a pet's health condition by analyzing data related to the pet's behavior. [Means for solving the problem]

[0006] According to the present disclosure, an information processing system is provided having a measurement data acquisition unit that acquires measurement data regarding a pet to be detected, a detection unit that detects changes in the pet's condition based on the acquired measurement data, and an output unit that outputs information corresponding to the detected changes in the pet's condition.

[0007] The present disclosure also provides a method including a processor acquiring measurement data relating to a pet to be detected, detecting a change in a state of the pet based on the acquired measurement data, and outputting information corresponding to the detected change in the state of the pet. Effect of the Invention

[0008] According to the present disclosure, it is possible to more accurately grasp the health condition of a pet. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram illustrating an example of a configuration of a system according to an embodiment of the present disclosure. [Diagram 2] FIG. 1 is a conceptual diagram illustrating sensing of animals in a home according to an embodiment of the present disclosure. [Diagram 3] FIG. 2 is a diagram illustrating an example of a hardware configuration of a server according to an embodiment of the present disclosure. [Figure 4] FIG. 2 is a diagram illustrating an example of a software configuration of a server according to the first embodiment of the present disclosure. [Diagram 5] 1 shows an example of the configuration of a measurement data acquisition unit and a measurement data storage unit according to the first and second embodiments of the present disclosure. [Figure 6] 4 illustrates configuration examples of an animal information generating unit and an animal information storing unit according to the first and second embodiments of the present disclosure. [Figure 7] FIG. 2 is an explanatory diagram of a flow of analysis of behavior types according to the first and second embodiments of the present disclosure. [Figure 8] FIG. 2 is a diagram illustrating input data analysis according to the first and second embodiments of the present disclosure. [Figure 9] FIG. 1 is an explanatory diagram of first and second embodiments of the present disclosure. [Figure 10] FIG. 2 is a diagram illustrating a method for measuring food intake according to the first and second embodiments of the present disclosure. [Figure 11] FIG. 2 is a diagram illustrating a method for measuring an excretion amount according to the first and second embodiments of the present disclosure. [Figure 12] FIG. 2 is a diagram illustrating a method for measuring body weight according to the first and second embodiments of the present disclosure. [Figure 13] FIG. 2 is a diagram illustrating a method for measuring food intake according to the first and second embodiments of the present disclosure. [Figure 14] FIG. 2 is a diagram illustrating an individual identification method according to the first and second embodiments of the present disclosure. [Figure 15] 2 is a configuration example of risk assessment information according to the first embodiment of the present disclosure. [Figure 16] 4 is a configuration example of notification determination information according to the first embodiment of the present disclosure. [Figure 17] FIG. 2 is a diagram showing a processing flow according to the first embodiment of the present disclosure. [Figure 18] FIG. 11 is a diagram illustrating an example of a software configuration of a server according to a second embodiment of the present disclosure. [Figure 19] FIG. 11 is a diagram illustrating a configuration example of proposed criteria information according to the second embodiment of the present disclosure. [Figure 20] FIG. 11 is a diagram illustrating a configuration example of solution information according to the second embodiment of the present disclosure. [Figure 21] FIG. 11 is a diagram illustrating a configuration example of determination information according to a second embodiment of the present disclosure. [Figure 22] FIG. 11 is a diagram illustrating an example of a screen display of a user terminal according to the second embodiment of the present disclosure. [Diagram 23] FIG. 11 is a diagram showing a processing flow according to a second embodiment of the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] The contents of the embodiment of the present disclosure will be listed and described below. One embodiment of the present disclosure has the following configuration. (Item 1-1) A measurement data acquisition unit that acquires measurement data relating to the pet to be detected; A detection unit that detects a change in the state of the pet based on the acquired measurement data; an output unit that outputs information corresponding to the detected change in the state of the pet; An information processing system having the above configuration. (Item 1-2) The information processing system described in item 1-1, wherein the detection unit detects a change in the condition of the pet based on animal information obtained by applying the animal information to a learning model created by machine learning based on the past measurement data of the pet to be detected. (Item 1-3) The information processing system according to item 1-2, wherein the detection unit detects a change in the pet's condition using at least one of quantitative information and / or a combination of qualitative information and quantitative information contained in the animal information. (Items 1-4) The information processing system according to any one of items 1-1 to 1-3, wherein the detection unit detects a change in a state of the pet based on information obtained from measurement data of a group corresponding to the pet to be detected. (Items 1-5) The information processing system according to any one of items 1-1 to 1-4, wherein the detection unit detects a change in a state of the pet based on information obtained from past measurement data of the pet to be detected. (Items 1-6) The detection unit detects a health risk of the pet based on a change in the state of the pet, The information processing system according to any one of items 1-1 to 1-5, wherein the output unit outputs information relating to the detected health risk of the pet. (Items 1-7) The information processing system described in item 1-6, wherein the output unit determines whether the detection status of the detected health risk satisfies a notification condition set according to the type of the health risk, and outputs if the notification condition is satisfied. (Items 1-8) The detection unit detects a change point of the measurement data, 8. The information processing system according to item 1-6 or 1-7, wherein the output unit outputs information regarding the time of the change point. (Items 1-9) A second measurement data acquisition unit is further provided for acquiring the measurement data of a second pet kept in the same environment as the pet to be detected, The information processing system described in any one of items 1-6 to 1-8, wherein the detection unit, when the health risk is detected, determines the likelihood of the health risk based on a comparison between the measurement data of the detection target and the measurement data of the second pet. (Items 1-10) The information processing system described in any one of items 1-1 to 1-9, wherein the output unit selects one or more solutions from a solution information storage unit that stores information about one or more solutions that can be proposed for changes in the pet's condition, and outputs information about the selected solutions. (Item 1-11) The information processing system described in items 1-10 further includes a determination unit that determines the effectiveness or utilization of the solution by comparing the measurement data acquired in a first period before the proposal with the measurement data acquired in a second period after the proposal. (Items 1-12) When the determination unit determines that the effect of the solution is low, Item 12. The information processing system according to item 1-11, wherein the output unit outputs information about a solution other than the solution. (Items 1-13) The information processing system according to any one of items 1-10 to 1-12, wherein the output unit selects a predetermined number of the solutions from one or more of the solutions that can be proposed according to priority, and outputs information about the selected solutions. (Items 1-14) The information processing system according to item 1-13, wherein the output unit preferentially selects solutions that have a proven track record of effectiveness in the past or solutions that have a proven track record of utilization. (Items 1-15) the measurement data is acceleration data; A behavior information generating unit calculates an index of a specific behavior from the acceleration data, The information processing system according to any one of items 1-1 to 1-14, wherein the detection unit detects a change in the state of the pet based on the behavior indicator. (Items 1-16) The processor: Obtaining measurement data relating to a pet to be detected; Detecting a change in the state of the pet based on the acquired measurement data; outputting information corresponding to the detected change in state of the pet; and A method comprising:

[0011] Another embodiment of the present disclosure has the following configuration. (Item 2-1) A measurement data acquisition unit that acquires measurement data relating to the pet to be detected; A risk assessment unit that detects health risks of the pet by applying the acquired measurement data to a learning model created by machine learning based on the past measurement data of the pet to be detected; A pet health abnormality notification system having a notification unit that notifies the detection result. (Item 2-2) The notification unit determines whether or not the detected detection status of the health risk satisfies a notification condition set according to a type of the health risk, The abnormal health notification system according to item 2-1, characterized in that notification is made when the notification conditions are met. (Item 2-3) The risk assessment unit detects a change point in the measurement data, The abnormal health notification system according to item 2-1 or 2-2, wherein the notification unit notifies information regarding the timing of the change point. (Item 2-4) A second measurement data acquisition unit is further provided for acquiring the measurement data of a second pet kept in the same environment as the pet to be detected, The health abnormality notification system described in any one of items 2-1 to 2-3, wherein when the health risk is detected, the risk assessment unit determines the likelihood of the health risk based on a comparison between the measurement data of the detection target and the measurement data of the second pet.

[0012] Another embodiment of the present disclosure has the following configuration. (Item 3-5) A measurement data acquisition unit for acquiring measurement data relating to a pet; A detection unit that detects a change in the state of the pet based on the acquired measurement data; and a solution proposal unit that selects one or more solutions from a solution information storage unit that stores one or more solutions that can be proposed for the change, and proposes the one or more solutions to a user. (Item 3-6) The information processing system described in item 3-1 further includes a determination unit that determines the effectiveness or utilization of the solution by comparing the measurement data acquired in a first period before the proposal with the measurement data acquired in a second period after the proposal. (Item 3-7) When the determination unit determines that the effect of the solution is low, The information processing system according to item 3-2, wherein the solution proposal unit proposes a solution different from the solution. (Item 3-8) The information processing system according to any one of items 3-1 to 3-3, wherein the solution proposal unit selects and proposes a predetermined number of the solutions in accordance with a priority from among one or more of the solutions that can be proposed. (Item 3-9) The information processing system according to item 3-4, wherein the solution proposal unit prioritizes the selection of solutions that have a high track record of effectiveness or utilization in the past. (Item 3-10) the measurement data is acceleration data; A behavior information generating unit calculates an index of a specific behavior from the acceleration data, The information processing system according to any one of items 3-1 to 3-5, wherein the detection unit detects a change in the state of the pet based on the behavior indicator.

[0013] <Details of the embodiment> Hereinafter, embodiments of the present disclosure will be described with reference to the drawings.

[0014] <Summary> A system according to an embodiment of the present disclosure monitors the health condition of a pet and detects abnormalities based on measurement data obtained from various sensors such as an acceleration sensor and a weight sensor 8 attached to the pet. As shown in Fig. 1, the system of the present disclosure includes a server 1, a pet sensor 5, a weight sensor 8, a communication terminal 2, and a user terminal 3, which are connected to the server 1 via a network such as the Internet. For convenience of explanation, Fig. 1 illustrates one each of the pet sensor 5, weight sensor 8, communication terminal 2, and user terminal 3, but multiple terminals of each can be connected to the network of this system.

[0015] The pet sensor 5 attached to the pet may be, for example, an acceleration sensor, a temperature sensor, etc. The weight sensor 8 may be one that measures the pet's weight, one that measures the pet's excrement (feces, urine), or one that measures the pet's intake (food, water), and may be provided on the bed, toilet, or bowl itself, or may be installed on a stand for these.

[0016] The data acquired by each sensor is transmitted to the server 1 via the communication terminal 2. The server 1 monitors the pet's condition by analyzing the acquired data, evaluates the health risk, and provides the user with necessary notifications. In the present disclosure, health risk means a high probability of contracting a particular disease, or a high probability of contracting the disease in the near future. In another embodiment, the server 1 monitors the pet's condition by analyzing the acquired data, detects changes in the health condition, and proposes necessary solutions. The proposed solution is evaluated for its effectiveness (whether the solution has improved the condition) and its degree of utilization (whether the solution is being utilized) by analyzing the measurement data acquired thereafter.

[0017] The server 1 can provide services to the user terminal 3 via an application. The user terminal 3 can transmit and receive information to and from the server 1 and can receive services by downloading the application from the server 1 or another server, executing the application, and accessing the server 1 via web page viewing software such as a browser.

[0018] The communication terminal 2 can acquire each data by performing short-distance wireless communication with the weight sensor 8 and the pet sensor 5 attached to an animal, for example, a cat 6. More specifically, first, as shown in FIG. 2, a collar-shaped (or pendant-shaped) wearable device is attached to an animal 6 such as a cat. The wearable device has an acceleration sensor and / or a temperature sensor built in. The weight sensor 8 and the pet sensor 5 transmit data to a receiving device 7 installed in the same house through short-distance wireless communication such as BLUETOOTH (registered trademark) LAW ENERGY (BLE), and the receiving device 7 transfers the data to a communication terminal 2 such as a router, and the communication terminal 2 transmits the data to a server 1 via a network. The weight sensor 8 and the pet sensor 5 may transmit data directly to the user terminal 3 through short-distance wireless communication such as BLUETOOTH (registered trademark) LAW ENERGY (BLE). Here, the receiving device 7 is equipped with a Linux (registered trademark)-based operating system, for example, and may also be equipped with various sensors such as a temperature sensor for measuring air temperature. However, the receiving device 7 may of course not be equipped with an OS, such as an embedded chipset.

[0019] The pet sensor 5 is, for example, an acceleration sensor. As shown in FIG. 2, the acceleration sensor 5 is a sensor that detects acceleration in three mutually orthogonal axial directions (x-axis, y-axis, and z-axis directions), and is built into a collar that is attached to the cat's neck. As shown in FIG. 2, the front-to-back direction of the cat is defined as the X direction, the left-to-right direction as the Y direction, and the up-to-down direction as the Z direction, and the collar is attached to the cat so that acceleration signals in each direction can be detected according to the cat's movements. The type of sensor is not limited to this, and any sensing device that can obtain information about the cat's movements, such as a gyro sensor or motion sensor, can be used.

[0020] The user terminal 3 may be, for example, a general-purpose computer such as a workstation or a personal computer, or may be a smartphone, a tablet, a mobile terminal, or another information terminal.

[0021] <Hardware configuration> The server 1 and the user terminal 3 according to this embodiment have the following hardware configuration: Note that the following configuration is an example, and they may have other configurations.

[0022] The server 1 is connected to a database (not shown) and constitutes a part of the system. The server 1 may be a general-purpose computer such as a workstation or a personal computer, or may be logically realized by cloud computing.

[0023] As shown in FIG. 3, the server 1 includes at least a control unit 10, a memory 11, a storage 12, a transmission / reception unit 13, an input / output unit 14, etc., which are electrically connected to one another via a bus 15.

[0024] The control unit 10 is a computing device that controls the operation of the entire server 1, controls the transmission and reception of data between each element, and performs information processing required for application execution and authentication processing, etc. For example, the control unit 10 is a CPU (Central Processing Unit), and executes programs stored in the storage 12 and deployed in the memory 11 to perform each information processing.

[0025] The memory 11 includes a main memory configured with a volatile storage device such as a DRAM (Dynamic Random Access Memory) and an auxiliary memory configured with a non-volatile storage device such as a flash memory, a HDD (Hard Disc Drive), etc. The memory 11 is used as a work area for the control unit (processor) 10, and also stores a BIOS (Basic Input / Output System) executed when the server 1 is started, various setting information, etc.

[0026] The storage 12 stores various programs such as application programs. A database (not shown) that stores data used in each process may be constructed in the storage 12.

[0027] The transmitting / receiving unit 13 connects the server 1 to a network. The transmitting / receiving unit 13 may include a short-range communication interface for Bluetooth (registered trademark) and Bluetooth Low Energy (BLE).

[0028] The input / output unit 14 includes information input devices such as a keyboard and a mouse, and output devices such as a display.

[0029] A bus 15 is commonly connected to each of the above elements, and transmits, for example, address signals, data signals, and various control signals.

[0030] <First embodiment> A first embodiment of the present disclosure will be described below. In this embodiment, the server 1 monitors the condition of the pet by analyzing the obtained data, evaluates the health risk, and provides the user with a necessary notification. FIG. 4 is a diagram showing an example of the software configuration of the server 1 in the system according to the first embodiment of the present disclosure. The server 1 can include a measurement data acquisition unit 21, an animal information generation unit 22, a risk assessment unit 23, a notification unit 24, a measurement data storage unit 31, an animal information storage unit 32, a user information storage unit 33, a risk information storage unit 34, and a notification determination information storage unit 35.

[0031] The measurement data acquisition unit 21, animal information generation unit 22, risk assessment unit 23, and notification unit 24 are realized by the control unit 10 provided in the server reading out and executing programs stored in storage 12 into memory 11, and the measurement data memory unit 31, animal information memory unit 32, user information memory unit 33, risk information memory unit 34, and notification judgment information memory unit 35 are realized as part of the memory area provided by at least one of the memory 11 and storage 12.

[0032] The measurement data acquisition unit 21 acquires data on the state of the animal from various measurement data acquisition means. As shown in FIG. 5, the measurement data acquisition unit 21 may include a behavior measurement data acquisition unit 211, a weight data acquisition unit 212, and an environmental data acquisition unit 213. The data on the state of the animal includes data on the behavior, activity, physical condition, and surrounding environment of the animal. The measurement data acquisition means here includes a pet sensor 5 such as an acceleration sensor, a weight sensor 8, a camera, and other devices that acquire data. In addition, various data acquired by the measurement data acquisition unit 21 is stored in the measurement data storage unit 31. As shown in FIG. 5, the measurement data storage unit 31 may include a behavior measurement data storage unit 311, a weight data storage unit 312, and an environmental data storage unit 313, and data acquired from the corresponding measurement data acquisition unit 21 is stored therein. The measurement data acquisition unit 21 and the measurement data storage unit 31 may include an acquisition unit and a storage unit that acquire other data such as body temperature data and image data of the animal in addition to these data.

[0033] The behavior measurement data acquisition unit 211 receives, via the transmission / reception unit 13 of the server 1, the behavior measurement data detected by the pet sensor 5 and transmitted via the communication terminal 2. The received behavior measurement data is stored in the behavior measurement data storage unit 311. Alternatively, the behavior measurement data can be stored in a storage built into the analysis server 9 shown in FIG. 1. The behavior measurement data is acceleration data when the pet sensor 5 is an acceleration sensor. Alternatively, the behavior measurement data may be image data.

[0034] The pet sensor 5 is, for example, an acceleration sensor. As shown in FIG. 2, the acceleration sensor 5 is a sensor that detects acceleration in three mutually orthogonal axial directions (x-axis, y-axis, and z-axis directions), and is built into a collar that is attached to the neck of an animal such as a cat. As shown in FIG. 2, the front-to-back direction of the animal is defined as the X direction, the left-to-right direction as the Y direction, and the up-to-down direction as the Z direction, and the collar is attached to the animal so that acceleration signals in each direction can be detected according to the animal's movements. The type of sensor is not limited to this, and any sensing device that can obtain information on the animal's movements, such as a gyro sensor or a motion sensor, can be used.

[0035] The weight data acquisition unit 212 acquires weight data from the weight measurement means. There are no particular limitations on the weight measurement means as long as it is a device equipped with a weight sensor 8 and capable of measuring the amount of food eaten, the amount of excretion, and the body weight. It is preferable that the weight measurement means has a shape (for example, a board type) on which pet items such as a toilet, a dish, and a water bowl can be placed depending on the purpose. The weight measurement means and the server 1 are connected by a communication network. It is preferable that the weight data is acquired in a time series. The acquired weight data is stored in the weight data storage unit 312 together with time information. The weight data storage unit 312 stores the weight data acquired by the weight data acquisition unit 212 for each weight measurement means. It is preferable that the weight data is time series data stored together with time data.

[0036] The environmental data acquisition unit 213 acquires environmental data in which the animals are kept. For example, the environmental data acquisition unit 213 acquires temperature data, room temperature data, humidity data, etc. obtained from thermometers and hygrometers installed inside and outside the room. Also, data on the climate of the area where the animals are located may be obtained from the Internet. Furthermore, the environmental data acquisition unit 213 acquires information related to disasters. Information related to disasters is information on occurrence of earthquakes, fires, floods, tsunamis, lightning strikes, tornadoes, etc. Information related to disasters may be acquired from various sensors installed inside and outside the room, or disaster information provided on the Internet may be acquired, and the method is not particularly limited. The acquired environmental data is stored in the environmental data storage unit 313.

[0037] The animal information generating unit 22 generates animal information related to the activity of the animal being measured by analyzing various measurement data. In other words, "animal information" is meaningful information (such as the weight of a specific measurement target, the number of times a specific behavior occurs, and the time) generated from the measurement data, which is raw data. As shown in FIG. 6, the animal information generating unit 22 can include a behavior information generating unit 221, a weight information generating unit 222, and the like. Furthermore, the data related to animals generated by the animal information generating unit 22 is stored in the animal information storage unit 32. The animal information storage unit 32 may include a behavior information storage unit 321 and a weight information storage unit 322.

[0038] The behavior information generation unit 221 generates behavior data of the animal based on the received behavior measurement data in cooperation with the analysis server 9 (or by independent processing in the behavior information generation unit 221).

[0039] Here, the behavioral information includes exercise data, sleep data, meal data, toilet data, position data, and the like stored in the behavioral information storage unit 321. More specifically, the exercise data includes aggregated data such as the amount of activity, such as whether or not exercise was performed, and how much activity was performed in a day over time; the sleep data includes aggregated data such as whether or not sleep was performed and how much sleep was performed in a day over time; the meal data includes aggregated data such as whether or not eating behavior was performed and how many meals were eaten and when they were eaten, aggregated data such as whether or not water intake behavior was performed and how many times water was drunk and when they were drunk; the toilet data includes aggregated data such as whether or not defecation behavior was performed and how many times defecation was performed and when they were performed, and aggregated data such as whether or not urination behavior was performed and how many times urination was performed and when they were performed. The position data includes the direction in which the animal moved and the location where it was located, and other data such as the time and number of times of grooming. In short, any information such as the time and number of times of behavior that can be labeled with a specific label may be used. Although not shown, the body temperature of the animal at the time of measurement may be measured.

[0040] Next, a generation flow of behavior information will be described. First, the behavior information generating unit 221 checks the measurement data detected by the behavior measurement data acquiring unit 211. Next, the behavior information generating unit 221 determines the behavior type based on the measurement data. The method of determining the behavior type can be realized by several known behavior analysis methods, but for example, the acceleration data (Gx, Gy, Gz) in the x, y, and z-axis directions obtained from the acceleration sensor 5 is decomposed into a period and an amplitude for each time using a wavelet transform, the periodicity of the signal at each time is recognized as a behavior spectrum, and the behavior can be classified according to the similarity of the spectrum by comparing it with the behavior elements registered in advance.

[0041] If there is no information on a behavioral element registered in advance, it can be recognized as a new behavioral element and provided to, for example, a veterinarian as data indicating abnormal behavior described later. Alternatively, for example, the acceleration data obtained from the acceleration sensor 5 can be Fourier transformed, and the average value or peak value of the frequency components calculated along the time axis can be compared with known frequencies corresponding to the behavioral types (exercise, sleep, eating, toilet, etc.) of the same or different animals to identify the behavior, or the acceleration components can be subjected to a fast Fourier transform (FFT) to extract characteristic waveforms or spectrum values ​​based on the calculated frequency components, and compared with known characteristic waveforms or spectrum values ​​corresponding to the behavioral types (exercise, sleep, eating, toilet, etc.) of the same or different animals to identify the behavior. In addition, the behavior type can be estimated by grasping the animal's posture from the posture in each axis direction (θx, θy, θz) calculated by the acceleration sensor 5.

[0042] When the behavior type is determined, the behavior information generating unit 221 generates data indicating the behavior type as behavior information together with the date and time when the measurement data was measured (or the date and time when the measurement data was received, or the date and time when the behavior information was generated).

[0043] Here, the flow of analysis of behavior types will be further explained with reference to FIG. 7. Acceleration data 101 acquired from the acceleration sensor 5 is subjected to data preprocessing 102 to convert it into spectrum data obtained by the above-mentioned wavelet transform or component data obtained by Fourier transform or the like. The data thus preprocessed is then scored 103 by a binary model group. The binary model according to this embodiment performs comparative analysis with models of activities that can be specifically expressed (interpreted), such as a WALK model, a RUN model, an EAT model, and a STAY model, and scores which behavior a specific part of the preprocessed 102 data can be inferred to be. For example, as shown in FIG. 8, the likelihood is scored by analyzing the input data with each model. In the illustrated example, "walking" is 91, "running" is 62, "eating" is 21, and "stopping" is 8, and the highest score is 91 for "walking", so the scoring result by the binary model group is classified as "walking".

[0044] Next, returning to FIG. 7, scoring 104 by the multi-value model group is performed. In the scoring by the multi-value model group according to this embodiment, when the results obtained by the binary models are competitive, it is determined based on machine learning which binary model group should be prioritized. For example, in the example shown in FIG. 8, the results are 91 for "walking" and 62 for "running", and the score of the evaluation of "running" is also relatively high. In this case, it is determined which binary model should be prioritized in this case based on the combination of the input data and the judgment results to the past binary model group. In this way, in this embodiment, the accuracy of the data is improved by further evaluating the results of the binary model group specialized in the judgment of each action by the multi-value model group.

[0045] Returning to Fig. 7, the determined behavior is further corrected based on the rule base. For example, if the binary model determines that a behavior that does not usually occur, such as "running," occurs suddenly during a determination interval of "eating" or "sleeping," which are behaviors that tend to continue for a certain period of time in cat behavior, or if the binary model cannot determine the behavior, the prediction result of the binary model for this interval is rejected, and a correction is made to estimate the behavior as another behavior based on the rules. When the correction is completed, the behavior is assigned 106, which is a behavior label that was registered in advance.

[0046] In this embodiment, feedback 107 is received from the user in order to respond to individual differences in each animal and individual factors due to the environment. Specifically, as shown in FIG. 9, the current behavior of the animal under one's care is recorded (manually) while observing it. By associating the record with the data from the acceleration sensor, teacher data can be collected by visual observation or the like. The feedback data 108 obtained in this way is accumulated and used to improve the accuracy of the models of the binary model group.

[0047] The behavioral information generated as described above is stored in the behavioral information storage unit 321. The behavioral information is preferably time-series data stored together with time data.

[0048] The weight information generating unit 222 analyzes the weight data acquired by the weight data acquiring unit 212 and calculates various measurement objects (body weight, food intake, water intake, feces amount, urine amount, etc.). The weight information generating unit 222 includes a weight calculating unit. The weight calculating unit analyzes the weight data acquired by the weight data acquiring unit 212 and outputs weight information of the target measurement object. If the weight acquiring means can set multiple measurement modes, the weight calculating unit calculates the weight according to the set measurement mode. The measurement mode defines at least the measurement object as described below, and can be set by the user, or may be automatically set by recognizing an item placed on the weight measuring means. An example of a method for calculating the weight for each measurement mode is shown below.

[0049] <Food intake, water intake, weight measurement mode> FIG. 10 shows an example of measuring the amount of food, water, and body weight. When an animal stands on the weight measuring means to eat or drink water, the time-series weight data behaves as shown in FIG. 10. The difference ΔW1 between the weight before the animal stands on the weight measuring means and the weight when the animal stands on the means can be regarded as the weight of the animal. The difference ΔW2 between the weight before the animal stands on the weight measuring means and the weight when the animal finishes eating and drinking and gets off the means can be regarded as the amount of food or water lost, that is, the amount of food or water consumed. If the weight data fluctuates due to the animal's movement while on the weight measuring means, it is acceptable to adopt an appropriate optimum value, such as the average value or the weight data when the animal has stopped moving for a certain period of time.

[0050] <Excretion amount / weight measurement mode> FIG. 11 shows an example of measuring the amount of excretion and body weight. When an animal enters the toilet above the weight measuring means to defecate or urinate, the time-series weight data will behave as shown in FIG. 11. The difference ΔW3 between the weight when the animal enters the toilet and the weight when the animal leaves the toilet can be considered as the weight of the animal. The weight can be the difference between the weight before the animal enters the toilet and the weight immediately after entering the toilet. Also, the difference ΔW4 between the weight before the animal enters the toilet and the weight after the animal leaves the toilet can be considered as the amount of excretion. If the weight data fluctuates due to the animal moving while it is in the toilet, it is acceptable to adopt an appropriate optimum value, such as an average value or weight data when there is no movement for a certain period of time.

[0051] <Weight measurement mode> Figure 12 shows an example of measuring weight. When an animal climbs onto a bed on the weight measuring means to sleep or relax, the time-series weight data will behave as shown in Figure 14. The difference ΔW5 between the weight when the animal climbs onto the bed and the weight when the animal climbs out of the bed can be regarded as the weight of the animal.

[0052] As described above, the weight calculation unit can estimate the weight of various measurement objects from the change in the weight data over time. The type of measurement object and the calculation method thereof are not limited to those described above, and can be set arbitrarily.

[0053] The weight information generating unit 222 may further include a weight information evaluating unit. The weight information evaluating unit compares the weight information calculated by the weight calculation unit with the behavior data to evaluate the accuracy of the weight information. The weight information evaluating unit refers to the behavior data at the time (t1 to t2) when the weight information of the measurement object is acquired from the behavior information storage unit 321, and checks whether the behavior of the pet at that time matches the measurement object of the weight information. For example, when the weight data fluctuates in the case of measurement in the food amount measurement mode, the amount of change in the weight data is judged as "food amount" as described above, but if the behavior data in that time period (t1 to t2) indicates "eating" as shown in FIG. 13, it can be judged that the weight information is likely to be the amount of food eaten. On the other hand, if the behavior data in that time period is not "eating" (for example, "playing"), it is judged that the weight information may not indicate the amount of food eaten. In this way, the weight information evaluating unit judges that the weight information is likely to be accurate when the behavior data in the same time period as the weight information matches the weight information, and can tag or delete the weight information as uncertain data when they do not match.

[0054] The weight information generating unit 222 may further include a weight type specifying unit. When the behavioral data can distinguish between eating rice and drinking water, or between defecation and urination, the weight type specifying unit can specify the weight information calculated by the weight calculating unit more specifically. For example, in the food amount / water intake / weight measurement mode, when both a dish and a water bowl are placed on the weight measuring means, ΔW2 indicates either the food amount or the water intake, or the total of both. Here, if the behavioral data from t1 to t2 indicates "eating", ΔW2 can be determined to be the food amount. Similarly, in the excretion amount / weight measurement mode, ΔW4 indicates either the defecation amount or the urination amount, or the total of both, but if the behavioral data in the time period in which the weight data was acquired was "defecation", ΔW4 can be determined to be the defecation amount. In this way, the weight type specifying unit can specify the measurement target of the weight information more specifically from the behavioral data in the same time period.

[0055] The weight information generating unit 222 may further include an individual identification unit. In the case of multiple animals, the individual identification unit can determine which individual the weight information calculated by the weight calculation unit is from. In the case of multiple animals, it is usually difficult to identify which individual is on one weight measurement means. The individual identification unit refers to the behavioral data of each individual during the time period (t1 to t2) when the weight data of the measurement target is acquired, and identifies the individual to which the weight information should be linked. In the example shown in FIG. 14, it is determined from the behavioral data of individuals A and B during t1 to t2 that the weight data acquired by the weight measurement means is that of individual A, who had a meal. In this way, the individual identification unit can select an individual showing behavioral data that matches each weight information, and assign individual information to the weight information.

[0056] Also, an individual may be identified from a waveform such as acceleration data obtained from a sensor of each individual. It is known that even for the same behavior, each individual has its own unique waveform characteristics. By comparing the characteristics of behavior waveform data during the time period when the weight data of the measurement subject was obtained with the waveforms of each individual's behavior registered in advance, an individual can be identified.

[0057] In addition to the above-mentioned methods, various methods may be used to identify the individual. For example, the individual may be identified by analyzing images obtained by an image acquisition means capable of photographing the weight measuring means. An image acquisition means such as a video camera can capture moving images over time, and the individual standing on the weight measuring means can be identified by image recognition. The individual identification unit can identify the individual to which the weight information should be linked from the image data at the time when the weight data of the measurement target is acquired.

[0058] The individual identification unit may identify an individual near the weight measurement means by receiving data including individual information from a pet's collar or the like, based on the strength of radio waves from BLUETOOTH (registered trademark) LAW ENERGY (BLE), etc. By providing a BLE receiving means on or near the weight measurement means, it is possible to recognize an individual that is closer to the weight measurement means.

[0059] The individual identification unit may identify an individual by weight. The weight of an individual is registered in advance, and when the weight is calculated in each measurement mode, the individual identification unit can identify the individual by referring to the registered weight.

[0060] As described above, the individual identification unit can identify individuals using a plurality of methods, and one or more of these methods can be adopted, or individual identification can be performed by combining a plurality of methods.

[0061] The weight information generated by the weight information generating unit 222 is stored in the weight information storage unit 322. For example, the measurement target and the weight of the measurement target (e.g., the amount of food eaten) may be stored for each measurement date and time. In addition, when multiple pets are kept, information on the individual names and individual IDs may be included.

[0062] The user information storage unit 33 manages basic information of the user (owner) and basic information of the animal kept by the user. The basic information of the user includes information such as gender, age, occupation, and address. The basic information of the animal kept by the user includes name, type (dog, cat, etc.), breed, age, gender, place of residence, keeping environment (outdoor, indoor), genetic information, whether or not multiple animals are kept, health information, and the like. Examples of health information include medical history and medical history. Information on multiple animals can be registered for one user. In addition, the basic information of the user and the basic information of the animal are managed by linking them to, for example, a user ID assigned to the user. An ID may also be assigned to the animal.

[0063] The risk assessment unit 23 assesses the health risk of the target animal based on the generated animal information. There may be multiple methods for risk assessment, and the risk assessment unit 23 may assess the risk using any one of the methods, or may perform a comprehensive assessment by combining the assessment results using multiple methods. The assessment methods will be described later. The risk assessment unit 23 is an example of a detection unit.

[0064] The server 1 in this embodiment may include a risk information storage unit 34 that stores the correspondence between animal information data and diseases. FIG. 15 is a configuration example of risk information stored in the risk information storage unit 34. The risk of a specific disease is associated with the type of animal information data and its change. Identification information such as an ID may be given to the disease, and identification information may also be appropriately given to the animal information data and the change. For example, an increase in the number of excretion times, an increase in the excretion time, and a decrease in the amount of movement are used as indicators of animal information for risk assessment of "cystitis". Similarly, an increase in the number of grooming times and an increase in the grooming time are used as indicators of animal information for risk assessment of "dermatitis", and a change in the amount of drinking water, a change in the number of excretion times, and a decrease in the amount of movement are used as indicators of risk assessment of "renal failure". In addition, a condition such as a predetermined threshold value may be included as a criterion for the change. The threshold value may be expressed as a numerical value. These combinations of disease and animal information are examples and can be changed as appropriate. For example, either the number of excretion times or the excretion time may be selected as an indicator in the risk assessment of "cystitis". The risk assessment unit 23 judges whether or not a change in each animal information data satisfies the conditions in the risk information storage unit 34 for determining that there is a risk of a specific disease.

[0065] The risk information storage unit 34 may include information on notification judgment conditions for judging whether or not to notify the user of each risk. FIG. 16 is an example of the notification judgment information. As shown in FIG. 16, a plurality of notification judgment conditions may be prepared. For example, it is possible to set the following: when any abnormal value is detected by the risk assessment unit 23, notification is always performed; when the detection of an abnormal value continues for a predetermined number of days or more; when the detection frequency of an abnormal value is a predetermined number or more; when an abnormal value is detected both by the analysis of the own data and by the comparative analysis of the data of other individuals, notification is performed; and the like. The notification judgment conditions are not limited to these, and can be set appropriately. In addition, as shown in FIG. 15, different notification conditions can be set for each disease. That is, appropriate notification criteria can be set according to the importance of the disease and the characteristics of the animal information, and it is possible to prevent the user from being unnecessarily anxious due to excessive notifications. In addition, when the degree of abnormality is judged in a plurality of stages, the notification conditions may be set to be different according to the degree of abnormality. For example, it is possible to set the notification condition B when the degree of abnormality is low, and the notification condition A when the degree of abnormality is high.

[0066] The notification unit 24 provides the risk assessment result performed by the risk assessment unit 23. The notification destination may be the user who is the owner, a veterinarian, a pet insurance company, or the like. The user may also register other users as notification destinations in advance. The other users may be, for example, family members or friends who watch over the pet as joint owners. In addition, conditions may be set for the notification so that excessive notifications are not made. In that case, it is confirmed that the notification conditions set in the notification judgment information for each disease are met, and the notification is made. The notification conditions may also be selectable by the user. The notification unit 24 is an example of an output unit. The manner of output by the notification unit 24 is not particularly limited, and may be, for example, a display on a terminal used by the user or doctor, output to a database of a server managed by a medical institution or a servicer, output by printing on a print medium such as paper, or output by voice.

[0067] A specific example of risk assessment will be described below.

[0068] <Disease prediction using supervised learning> The risk assessment unit 23 may perform disease prediction by machine learning. A learning model may be created using the past animal information data of an animal with a clear disease history stored in the animal information storage unit 32 as input data and the disease data of the animal as teacher data. The risk assessment unit 23 applies the animal information data of the evaluation target animal to the created learning model to evaluate the possibility of the animal contracting a specific disease. There is no particular limit to the animal information data used for prediction, but it is preferable to use one or more pieces of data related to diseases, such as the number of excretion times, time, and amount of excretion, activity level, number of meals and amount, and number of grooming times and time. For example, when the number of excretion times and the amount of exercise are used to predict cystitis, the input data may be animal information on the number of excretion times and the amount of exercise of an animal with a record of contracting cystitis just before the animal contracted the disease. In addition, when evaluating the risk of cystitis of an evaluation target individual, the animal information on the number of excretion times and the amount of exercise of the animal may be applied to the learning model to evaluate the risk of cystitis. In addition, when the number of grooming episodes and the grooming time are used to predict dermatitis, the animal information of an animal with a record of having dermatitis, including the number of grooming episodes and the grooming time immediately before the onset of the disease, can be used as input data. In addition, when evaluating the risk of dermatitis in an individual to be evaluated, the number of grooming episodes and the grooming time of the animal can be applied to the learning model to evaluate the risk of dermatitis.

[0069] <Detecting anomalies by comparing with own data> The risk assessment unit 23 may assess health risks by analyzing the animal information of the individual to be assessed and detecting abnormal values. Since animals vary greatly from one another, it is useful to perform risk assessment by capturing changes in the individual's data. The risk assessment unit 23 reads out the animal information of the individual to be assessed from the animal information storage unit 32 and performs a predetermined process to assess the risk. For example, the individual may be assessed as being at risk when the difference from the average value in a predetermined period of time in the past is equal to or greater than a predetermined rate. Specifically, the risk assessment unit 23 may detect abnormalities in the individual by comparing past animal information (or measurement data) with current animal information (or measurement data) for the same individual. More specifically, when the average time of the running state in the animal information last week is 10 minutes, while the average time of the running state in the animal information this week is 82 minutes, the risk assessment unit 23 may detect that the individual is not in good spirits, and the notification unit 24 may notify that the individual is not in good spirits. In this case, the threshold value for whether or not to notify may be determined, for example, for each individual, or for each type of individual. The notification frequency of the abnormality detection by the notification unit 24 may be configurable as necessary. For example, since the amount of activity decreases during illness, notification regarding abnormalities may not be made for a certain period of time. For example, the notification unit 24 may adjust the frequency and manner of notification based on the history of the above-mentioned animal information.

[0070] The risk assessment unit 23 may also determine abnormal values ​​by a statistical method. That is, an estimated value is calculated by an autoregressive model based on the animal information data, which is time-series data, and when the actual animal information data is significantly different, it can be determined that the data deviates from the past data series. The autoregressive model can be an AR model, an MA model, an ARMA model, an MRIMA model, or any other known model that matches the characteristics of the data.

[0071] In addition, the Mahalanobis distance is calculated based on the animal information data, which is time-series data, and if the Mahalanobis distance is large, it can be determined that the data is deviated from a set of past data series. A known method can be used to calculate the Mahalanobis distance. In addition, weighted data that has been weighted so that the weight is relatively increased according to the newness of the time-series data may be used for the past time-series data.

[0072] The risk assessment unit 23 may detect a change point indicating the time when the data started to change by statistical processing. That is, the risk assessment unit 23 searches for data that can be recognized as a sign of an abnormality before the abnormality threshold is exceeded. By detecting a change point, it is possible to know when the change started to appear, which is useful for the user to search for the cause. The detection of abnormal values ​​by analyzing the own data is not limited to the above method, and any method known as an outlier detection or change point detection method can be appropriately adopted.

[0073] <Detecting abnormalities by comparing data from other individuals> The risk assessment unit 23 may detect an abnormal value by comparing the animal information data of the individual to be evaluated with the animal information data of an individual other than the individual to be evaluated. The group of other individuals to be compared may be similar individuals that have at least one attribute in common with the individual to be evaluated, such as type (dog, cat, etc.), breed, age, sex, place of residence, rearing environment (outdoor, indoor), genetic information, whether multiple animals are kept, and health information. In addition, these attributes may be "similar" with a certain range. For example, if the individual to be evaluated is type = dog, breed = toy poodle, age = 3 years, and sex = male, an individual corresponding to type = dog, breed = small dog, age = 2 to 4 years, and sex = male may be set as a similar individual, and an appropriate range may be set. In addition, a group means a group that is classified according to attributes such as age and type of animals. In this way, when a certain individual is detected by comparison to be in a state different from a healthy state in a normal group, the notification unit 24 may notify that the individual is abnormal. The notification unit 24 may also determine whether or not to notify based on a comparison with the animal information of the group. For example, seasonality is observed in the behavior of animals. Therefore, if only the past animal information of an individual is used as the comparison target, a risk of abnormality based on an increase or decrease in a specific behavior may be notified even if the animal is healthy. Therefore, by comparing not only past and current animal information but also with the animal information of the group, risk can be evaluated with higher accuracy.

[0074] The risk assessment unit 23 compares the animal information data of other individuals with the animal information data of the evaluation target animal to assess the health risk of the evaluation target animal. For example, the risk assessment unit 23 may detect an abnormality when the difference between the average value or rate of change of the animal information data in a predetermined period of time and the average value or rate of change in a group of other individuals is equal to or greater than a predetermined value (ratio). In addition, the risk assessment unit 23 may detect an abnormality when the probability distribution in the group of other individuals falls within a predetermined range between upper and lower. The method of comparing the data of other individuals is not limited to the above, and any known method may be adopted as appropriate.

[0075] In any of the above methods, the criteria for determining the health risk may be set in multiple stages. For example, multiple thresholds may be set as criteria to determine the degree of abnormality (low, medium, high).

[0076] In the above method, which of the acquired animal information data is used for risk assessment can be selected according to the relationship between the animal information data registered in the risk information storage unit 34 and the disease. Regarding behavioral changes that appear as signs of an animal's disease, for example, a change in the amount of water consumed, a change in the number of times the animal goes to the toilet, or a decrease in the amount of exercise may indicate renal failure, an increase in the number of times the animal goes to the toilet may indicate diabetes, an increase in the number or time of going to the toilet or a decrease in the amount of exercise may indicate cystitis, and an increase in the number or time of grooming may indicate dermatitis. According to this system, data on the number, time, and amount of behaviors such as excretion, eating, amount of exercise, and grooming can be acquired, so that risk can be assessed by selecting and analyzing the relevant animal information data according to the disease risk to be known.

[0077] FIG. 17 is an operation flow of the server 1 according to the embodiment of the present disclosure.

[0078] First, the measurement data acquisition unit acquires measurement data on the animal to be evaluated (S301). The measurement data is various data obtained from an acceleration sensor, a temperature sensor, a weight sensor for measuring the amount of food eaten, the amount of excretion, and the weight of the animal, etc. The acquired measurement data is stored in the measurement data storage unit 31.

[0079] The animal information generating unit 22 generates animal information based on the obtained measurement data (S302). While the measurement data is raw data obtained from the sensor, the animal information is information obtained by analyzing the raw data, and is information such as the number and time of each behavior (eating, excretion, playing, sleeping, walking, grooming, etc.), weight data of meals and excretion, weight, etc. The generated animal information data is stored in the animal information storage unit 32.

[0080] The risk assessment unit 23 assesses the health risk of the target animal based on the generated animal information. Any of the assessment methods described above may be adopted, or other known methods may be adopted. The risk assessment unit 23 judges whether the detected change in the animal information corresponds to a specific disease risk based on the risk assessment information (S303). If a change that satisfies the conditions of the risk assessment information is detected, it is determined that there is a risk of the disease (S304=Yes).

[0081] When the risk assessment unit 23 determines that there is a disease risk, the notification unit 24 determines whether the notification conditions stored in the notification determination information storage unit 35 are satisfied (S304). When the notification conditions differ for each disease, the notification unit 24 reads out the notification conditions set for the target disease from the notification determination information storage unit 35 and determines whether the predetermined notification conditions are satisfied. When it is determined that a notification should be made, the output unit performs the notification by outputting the risk information to a predetermined notification destination such as a user terminal (S305).

[0082] According to the present disclosure, it is possible to detect changes in the life of a pet and notify a user of health risks of the pet. Also, according to the present disclosure, it is possible to obtain a wide range of information on the behavior of a pet, thereby making it possible to make a highly accurate judgment on the risk of disease based on multiple pieces of information. Also, it is possible to suppress the adverse effects of too many notifications by flexibly setting notification conditions for notifying a user of detected abnormalities.

[0083] (Variation 1) This modification assumes the case of a user who keeps multiple pets. This modification judges the likelihood of a health risk detected in the evaluation target individual based on data on multiple pets kept in the same environment as the evaluation target individual.

[0084] Hereinafter, the individual to be evaluated is referred to as a first animal, the data of that animal is referred to as first animal information, another individual raised in the same environment as the individual to be evaluated is referred to as a second animal, and the data of that second animal is referred to as second animal information. The measurement data acquisition unit 21 and the animal information generation unit 22 process the second animal in the same manner as the first animal, and continuously generate various types of animal information.

[0085] The risk assessment unit 23 assesses the health risk of the first animal, which is the individual to be assessed, based on the first animal information. Then, when any risk is detected, the second animal information during the period in which the abnormality was detected is read and compared. Then, it is confirmed whether the second animal information also shows a similar tendency. That is, if the second animal information also shows a tendency of decreased food intake during a period in which a significant decrease in food intake was observed in the first animal information, it is highly likely that other factors due to the rearing environment are involved rather than a sign of the first animal's health deterioration. For example, it is possible that the change in food has decreased the appetite of the two animals, that the room temperature is too high and they have lost their appetite, or that the environment around the dishes is different from usual (a new object that makes noise has been placed near the dishes, etc.). In addition, if the activity level of both animals increases, it is possible that the cause is excitement due to a change in the environment, such as the introduction of a new toy or the arrival of a visitor. In this way, by comparing data from multiple animals and checking whether there is a similar tendency, it is possible to determine that the animal information is showing an abnormal value due to a change in the rearing environment, even if the cause is not necessarily clear, that is, the certainty of the detected health risk is low.

[0086] If a similar tendency is found as a result of the comparison with the second animal information, the risk assessment unit 23 determines that there is no abnormality in health and does not issue a notification. Alternatively, when notifying that an abnormality has been detected, the notification may be accompanied by information that the abnormality may be due to the rearing environment.

[0087] (Variation 2) In this modification, a modification of the method for detecting an anomaly by the risk assessment unit 23 will be described.

[0088] For example, the risk assessment unit 23 may detect an abnormality in an animal based on a comparison between the animal information or the measurement data and a threshold value based on a specific symptom. The risk assessment unit 23 may also detect an abnormality in an animal based on a result of an appropriate combination of the above-mentioned comparison with past animal information, comparison with the animal information of the group, and comparison with a threshold value. For example, the risk assessment unit 23 may detect a change in the state of an animal by using the statistical distribution of the past animal information of an individual and the statistical distribution of the past animal information of the group or the whole. Specifically, when there is a difference between the statistical distribution of the animal information of the group or the whole (i.e., a general distribution) and the statistical distribution of the animal information of an individual, a threshold value specialized for the individual may be set using an integrated statistical distribution obtained by weighting each statistical distribution. This can improve the accuracy of risk assessment.

[0089] The information (parameters) included as the animal information include, for example, quantitative information such as urine volume, body weight, and feces volume, and information (qualitative information) regarding qualitative activities such as drinking water, eating, and walking. When detecting a change in the animal's condition using a threshold value, for example, for the qualitative information, a threshold value can be set by combining it with quantitative information such as the number (frequency) and time of such activities. That is, the risk assessment unit 23 can detect a change in the animal's condition based on such quantitative information, or a combination of qualitative information and quantitative information. For example, the risk assessment unit 23 may detect the possibility of some abnormality in the animal's thyroid gland based on the fact that the urine weight has increased to a predetermined rate or more and the animal's walking or running time tends to increase.

[0090] Furthermore, the size of the span acquired as the animal information is not particularly limited. For example, the span of the acquired animal information may be appropriately determined in units of seconds, minutes, hours, days, weeks, months, etc. These spans may be appropriately determined according to the degree of certainty required for each parameter included in the animal information (i.e., the length of time required to detect a change in the animal's condition). In addition, in a relatively short span, the risk assessment unit 23 may be able to detect abnormal behavior itself, such as convulsions or coughing.

[0091] <Second embodiment> The system according to the second embodiment of the present disclosure monitors the condition of a pet based on measurement data obtained from various sensors such as an acceleration sensor and a weight sensor 8 attached to the pet, and proposes a necessary solution. FIG. 18 is a diagram showing an example of the software configuration of a server 1 in the system according to the second embodiment of the present disclosure. The server 1 can include a measurement data acquisition unit 21, an animal information generation unit 22, a detection unit 25, a solution selection unit 26, an output unit 27, a judgment unit 28, a measurement data storage unit 31, an animal information storage unit 32, a user information storage unit 33, a proposal reference information storage unit 36, a solution information storage unit 37, and a judgment information storage unit 38.

[0092] The measurement data acquisition unit 21, animal information generation unit 22, detection unit 25, solution selection unit 26, output unit 27, and judgment unit 28 are realized by the control unit 10 of the server reading out and executing a program stored in the storage 12 into the memory 11, and the measurement data memory unit 31, animal information memory unit 32, user information memory unit 33, proposed standard information memory unit 36, solution information memory unit 37, and judgment information memory unit 38 are realized as part of the memory area provided by at least one of the memory 11 and the storage 12.

[0093] Here, the functions possessed by each of the measurement data acquisition unit 21, the animal information generation unit 22, the measurement data storage unit 31 and the animal information storage unit 32 are similar to the functions possessed by each functional unit in the first embodiment, and therefore will not be described.

[0094] The user information storage unit 33 manages basic information of the user (owner) and basic information of the animal kept by the user. The basic information of the user includes information such as gender, age, occupation, and address. The basic information of the animal kept by the user includes name, type (dog, cat, etc.), breed, age, gender, place of residence, keeping environment (outdoor, indoor), genetic information, whether or not multiple animals are kept, health information, and the like. Examples of health information include medical history and medical history. Information on multiple animals can be registered for one user. In addition, the basic information of the user and the basic information of the animal are managed by linking them to, for example, a user ID assigned to the user. An ID may also be assigned to the animal.

[0095] Also, information on the priority of solutions can be stored as basic information of the user. The priority of solutions is the order of priority selection for the types of solutions proposed in this system. For example, in the case where the types of solutions include the introduction of goods, the introduction of food, the use of medicine, the use of supplements, etc., if it is desired to avoid the use of medicine or supplements as much as possible, the priority of these is set low. The user information storage unit 33 may store information on the priority of each solution, or may store information on solutions that should be particularly avoided or selected.

[0096] The proposed criterion information storage unit 36 ​​can set criteria for proposing a solution for the state and changes of each index in the animal information (body weight, number of times, amount, and time of eating and excretion, activity level, number and time of various behaviors such as grooming, playing, and sleeping, etc.). FIG. 19 shows an example of the configuration of proposed criterion information stored in the proposed criterion information storage unit 36. The proposed criterion information is stored with each criterion linked to an ID. For example, if the average body weight in the past week has increased by 10% compared to the average value in a predetermined period of the past, "weight increase" (ID: 001) is detected. The setting of the criteria shown in FIG. 19 is merely an example, and can be set appropriately. In addition to the change in the target individual itself, a criterion compared to other individuals may be set. For example, individuals that share one or more common attributes, such as animal species (dog, cat, etc.), breed (American shorthair, Abyssinian, mixed breed, etc. for cats), age, sex, genetic information, presence or absence of chronic illness, type of chronic illness, rearing environment (outdoor, indoor), presence or absence of multiple pets, etc., may be grouped to form other individual data, and a statistical value compared with the data of the other individuals may be used as the criterion. In other words, a criterion may be set such as that the value of a certain index is in the top 10% of other individuals, or that the deviation from the average value of other individuals is a predetermined value or more. The criterion may be set based on one index or may be defined based on multiple indexes. For example, a new index may be calculated based on multiple data, such as calculating the calorie consumption of an animal from its weight and activity level, and using the calculated increase or decrease in calorie consumption or the difference between calorie intake and calorie consumption as the criterion. The type of index and calculation formula used to calculate the calorie consumption may be selected appropriately.

[0097] The solution information storage unit 37 determines and stores what kind of solution to provide when it detects a state or change in the animal information (weight, number of times, amount, and time of eating and excretion, activity level, number of times and time of various behaviors such as grooming, playing, and sleeping, etc.). FIG. 20 shows an example of the configuration of information stored in the solution information storage unit 37. One or more solutions are associated with each state or change in the animal information. The solution may be anything that brings about some change in the life of the animal, such as the introduction of goods, the intake of medicine or supplements, a change in the diet, or a change in the rearing environment. For example, the solution may be, for weight gain, the introduction of exercise equipment to increase the amount of exercise, the introduction of diet food, the use of dermatitis supplements or the use of an Elizabethan collar since dermatitis is suspected when the number of times or time of grooming has increased, the improvement of the toilet environment since cystitis is suspected when the number of times or time of toileting has increased, the increase of the number of times of toileting is suspected, and the increase of the number of water bowls to increase water intake. These are merely examples and are not limited. Each solution may be assigned a solution ID. The solution ID may be assigned as a combination of the state or change in the animal information and the content of the solution, for example, as "change ID + solution ID." For example, in the example of Fig. 20, "001001" is assigned to "introduce exercise equipment," which is the solution for change ID: 001. The solution ID assignment rules are merely examples, and are not limited to these.

[0098] Classification information by type may be added to each solution. Examples of classification information include, but are not limited to, the introduction of exercise equipment or the use of an Elizabethan collar as "goods", the introduction of diet food as "food", the use of dermatitis supplements as "medicine / supplement", etc. Classification information is useful for determining the priority of solutions according to user preferences. For example, a user who wants to avoid using medicines and supplements can lower the priority of solutions that have the classification information of "medicine / supplement" attached. Classification information may also be attached by a classification ID.

[0099] Also, information on past performance of effectiveness and utilization can be added to each solution. Past performance of effectiveness and utilization is an index showing how effective and utilized the solution was when it was selected in the past, and may be performance on other animals or only on the target animal itself. By adding past performance of effectiveness and utilization, the priority order of solutions can be adjusted.

[0100] The past track record of utilization and effectiveness can be calculated, for example, using the results of effectiveness and utilization assessments for animals to which the same solution was proposed in the past. When calculating the track record based on the presence or absence of effectiveness and utilization, the track record may be calculated as "the number of times that effectiveness and utilization were 'present' / the number of past proposals." For example, if the number of times a solution was proposed in the past was 100, of which 30 were judged to be effective and 50 were judged to be utilized, the track record of effectiveness is 30% and the track record of utilization is 50%. In addition, when only data on the target individual animal is used as the track record, the track record may be calculated as "the number of times that effectiveness and utilization were 'present' / the number of past proposals" for the past proposed solutions for the target individual.

[0101] The judgment information storage unit 38 includes information that is a criterion for judging how much each solution is utilized and how effective it is. FIG. 21 shows an example of the configuration of judgment information stored in the judgment information storage unit 38. For each solution, either or both of the utilization degree judgment criterion and the effect judgment criterion are set. The utilization degree judgment criterion and the effect judgment criterion can basically be defined by the presence or absence of a change in the animal information or the amount of change. For example, the utilization degree judgment criterion for a solution such as the introduction of exercise equipment can be determined by a certain increase in the amount of activity of the animal, specifically, "the average amount of activity per day after the introduction has increased by 10% or more compared to the average amount before the introduction". In addition, the effect judgment criterion for a solution for weight gain can be determined by a certain decrease in weight, specifically, "the weight after the introduction has decreased by 5% or more compared to the weight before the introduction". The judgment criterion may be defined by a reference numerical value (for example, the amount of change (difference) from the original numerical value or the value of the change ratio), or may simply be defined by an increase or decrease in a numerical value. In addition, if it is difficult or unnecessary to assess the degree of utilization depending on the solution, it is possible to set no utilization assessment criteria. For example, it is possible to set no utilization assessment criteria for solutions where the degree of utilization is difficult to understand from measurement data, such as the use of supplements, or for solutions where the use of an Elizabethan collar or improvements to the toilet environment are considered to be utilized if the owner, the user, implements them.

[0102] The detection unit 25 detects changes in the life of the animal from the animal information. In the detection, the criteria for proposing a solution for each index (weight, number of times, amount, and time of eating and excretion, activity level, number and time of various behaviors such as grooming, playing, and sleeping, etc.) are set in advance in the proposal criteria information storage unit 36. The detection unit 25 applies the obtained animal information to the proposal criteria information to check whether there is any that satisfies the criteria. For example, when the detection unit 25 obtains weight information on a certain day, it calculates the average weight for the past week including that day and compares it with the average weight for a predetermined period in the past. If the most recent average weight has increased by 10% from the past average weight, the detection unit 25 detects "weight gain" (ID: 001) and outputs the detected information to the solution selection unit 26 to propose a solution for the weight gain.

[0103] The detection unit 25 can also detect changes by an anomaly detection algorithm based on machine learning. In this case, the detection unit 25 can detect abnormal values ​​for predetermined movement information by loading newly obtained data into a learning model that has been machine-learned to learn time-series data for each measurement data (animal information) of the animal to be measured.

[0104] The solution selection unit 26 selects a solution to be proposed to the user based on the change in the animal information detected by the detection unit 25. The solution selection unit 26 reads out a solution associated with the ID of the detected change from the solution information storage unit 37. For example, when the change ID is 001, two solutions are set: introduction of exercise equipment and introduction of diet food. The solution selection unit 26 may select all solutions linked to the change ID, or may further select a predetermined number from among them. The solution selection unit 26 transmits the solution ID of the selected solution to the output unit 27.

[0105] The solution selection unit 26 may select solutions not exceeding a predetermined number. In that case, the solutions selected are ordered from those having a higher priority among those having a higher priority. The priority can be set based on the user's preference for the type of solution registered in advance in the user's basic information, or on the past effectiveness evaluation results and utilization evaluation results. When the solution selection unit 26 is to determine the solution based on the user's preference, the solution selection unit 26 reads out preference information linked to the user ID from the user information storage unit 33, and selects a solution type that matches the preferred solution type. When the solution selection unit 26 is to select the solution based on the past effectiveness evaluation results and utilization evaluation results, the solution selection unit 26 gives priority to the solution with a good past effectiveness evaluation result or utilization evaluation result recorded for each solution.

[0106] The output unit 27 outputs the contents of the solution to the user terminal 3 based on the selected solution ID. FIG. 22 is an example of a screen display of the user terminal 3. The contents to be output include information on the detected change, an outline of the solution, past effectiveness assessment results, utilization assessment results, a link to an EC site, etc. (in the case of a solution related to the purchase of goods or food). The output unit 27 may also be provided with an input button for requesting the proposal of another solution. When a request for another solution is input, the solution selection unit 26 selects a different solution, and the output unit 27 displays it. For example, the solution selection unit 26 selects a solution with the next highest priority after the previously proposed solution.

[0107] After proposing a solution, the determination unit 28 determines whether the solution has been implemented (degree of utilization) and whether it has been effective. In making the determination, an input may be received from the user that the implementation of the proposed solution has started. Alternatively, the start of implementation may be determined based on the user's purchase of the proposed goods, food, medicine, etc. on an EC site, or the proposed goods, etc. being recognized in an image taken of a pet's environment.

[0108] The judgment unit 28 analyzes the data acquired by the measurement data acquisition unit 21 after the solution is proposed or after the implementation is started, and makes a judgment based on the utilization judgment criteria or the effectiveness judgment criteria set in the judgment information storage unit 38. If the utilization judgment criteria are met, it is judged that the animal has fully utilized the solution, and if the effectiveness judgment criteria are met, it is judged that the solution has been effective. The judgment may be made not only based on the presence or absence of utilization or the presence or absence of effectiveness, but also by judging the degree of utilization or the degree of effectiveness numerically. The judgment result is stored in a database such as the solution information storage unit 37 or the user information storage unit 33, and is output to the user terminal 3 by the output unit 27.

[0109] If the result of the utilization degree judgment or the effectiveness judgment is not good, the solution selection unit 26 may select another solution and propose it to the user again. In that case, the proposed solution is excluded from the options, and another solution with the next highest priority is selected.

[0110] FIG. 23 is an operation flow of the server 1 according to the embodiment of the present disclosure.

[0111] First, the measurement data acquisition unit 21 acquires various measurement data on a daily basis (S401). The animal information generation unit 22 generates movement information such as weight, number of meals, amount, and duration of excretion, activity level, and number and duration of various behaviors such as grooming, playing, and sleeping based on the measurement data (S402).

[0112] The detection unit 25 checks whether any change has occurred for which a solution should be proposed, based on the motion information data (S403). Specifically, the detection unit 25 detects a change that satisfies the criteria registered in the proposal criteria information storage unit 36, together with its change ID. If a change is detected (S403=Yes), the detection unit 25 transmits the detected change ID to the solution selection unit 26.

[0113] The solution selection unit 26 reads out the solutions corresponding to the detected change from among the solutions set in the solution information storage unit 37 (S404). If the number of solutions to be proposed is fixed, a predetermined number of solutions are selected based on priority. The priority can be set based on whether the solution matches the user's preference regarding classification of solutions or whether the solution has a good track record of past utilization and effectiveness. The solution selection unit 26 transmits the selected solution ID to the output unit 27, and the output unit 27 outputs information related to the selected solution to the user terminal 3 (S405).

[0114] After the solution is proposed, the measurement data acquisition unit 21 acquires the measurement data and generates the animal information again (S406, S407). Note that it is preferable that the measurement data acquisition unit 21 and the animal information generation unit 22 continuously acquire and analyze data regardless of whether a solution is proposed or not.

[0115] The judgment unit 28 judges the effectiveness and the utilization level based on the data acquired after the solution is proposed (S408). The judgment unit 28 reads out the criteria for judging the effectiveness and the utilization level of the implemented solution from the judgment information storage unit 38, and makes a judgment according to the criteria. In addition, the judgment period for the effectiveness and utilization level of the solution can be set to the period from the time when the notification that the implementation of the solution has started is received from the user.

[0116] The result of the determination is recorded in a storage unit such as the solution information storage unit 37 or the user information storage unit 33 (S409). Data on the effectiveness and whether or not each solution was utilized is accumulated, and can be used to prioritize solutions when selecting them. If the result of the determination is that the solution did not have the predetermined effect or was not utilized (S408=No), the solution selection unit 26 can propose a solution different from the one previously selected (S404).

[0117] According to the present disclosure, it is possible to detect changes in the life of a pet, propose an appropriate solution to the user who is the pet's owner, and verify the effectiveness of the proposed solution based on the change in data. In addition, if the owner is often absent, it is difficult to grasp how much the pet is utilizing the solution provided to the pet, but the degree of utilization can be determined by analyzing the data.

[0118] In addition, by selecting the next solution based on past usage and effectiveness, it becomes possible to propose a solution suited to the preferences and characteristics of the target individual.

[0119] The functions of the modified comparison of animal information described in Modification 2 of the first embodiment, detection based on the type of animal information, and the span of the animal information to be compared can also be performed by the detection unit 25 of this embodiment.

[0120] The above-described embodiment is merely an example for facilitating understanding of the present disclosure, and is not intended to limit the present disclosure. The present disclosure can be modified or improved without departing from the spirit of the present disclosure, and it goes without saying that the present disclosure includes equivalents thereof.

[0121] In addition, the techniques described in each of the above embodiments can be combined with each other. For example, based on information on a change point in the pet's condition detected by the risk assessment unit 23, the solution selection unit 26 may select a solution suitable for the information on such a change point. In addition, the risk assessment unit 23 may evaluate health risks based on information on the utilization degree and effect of the solution selected by the solution selection unit 26, information on the history, etc. In addition, the risk assessment unit 23, the notification unit 24, the detection unit 25, the solution selection unit 26, the output unit 27, and the determination unit 28 may perform processing based on information obtained by the functional units according to other embodiments. [Explanation of symbols]

[0122] 1 Server 2. Communication terminals 3. User terminal 5 Pet Sensor 8 Weight Sensor 21 Measurement data acquisition section 22 Animal information generation section 23 Risk Assessment Division 24 Notification Department 25 Detection unit 26 Solution Selection Section 27 Output section 28 Judgment section 31 Measurement data storage unit 32 Animal information storage section 33 User information storage unit 34 Risk information storage unit 35 Notification judgment information storage unit 36 Proposed standard information storage unit 37 Solution information storage unit 38 Judgment information storage unit

Claims

1. A measurement data acquisition unit that acquires measurement data relating to the pet to be detected; a detection unit that detects the number of times a specific behavior occurs and the time taken for the behavior based on the acquired measurement data, and detects a health risk of the pet based on at least one of an increase and a decrease in the number of times the behavior occurs and at least one of an increase and a decrease in the time taken for the behavior; an output unit that outputs information regarding the detected health risk of the pet; An information processing system having the above configuration.

2. A measurement data acquisition unit that acquires measurement data relating to the pet to be detected; a detection unit that detects the number of times a specific behavior occurs and the amount of movement based on the acquired measurement data, and detects a health risk of the pet based on at least one of an increase and a decrease in the number of times the behavior occurs and at least one of an increase and a decrease in the amount of movement; an output unit that outputs information regarding the detected health risk of the pet; An information processing system having the above configuration.

3. The information processing system according to claim 1 , wherein the detection unit generates the number of times the behavior and the time taken for the behavior by applying the past measurement data of the pet to be detected to a learning model created by machine learning.

4. The information processing system of claim 2, wherein the detection unit generates the number of times the behavior occurs and the amount of movement by applying the past measurement data of the pet to be detected to a learning model created by machine learning.

5. The information processing system according to claim 1 or 2, wherein the output unit determines whether the detection status of the detected health risk satisfies a notification condition set according to the type of the health risk, and outputs the detection status if the notification condition is satisfied.

6. The detection unit detects a change point of the measurement data, The information processing system according to claim 1 , wherein the output unit outputs information relating to a time of the change point.

7. A second measurement data acquisition unit is further provided for acquiring the measurement data of a second pet kept in the same environment as the pet to be detected, The information processing system of claim 1 or 2, wherein the detection unit, when the health risk is detected, determines the likelihood of the health risk based on a comparison between the measurement data of the detection target and the measurement data of the second pet.

8. The information processing system of claim 1, wherein the output unit selects one or more solutions from a solution information storage unit that stores information regarding one or more solutions that can be proposed for at least one increase and decrease in at least one of the number of times the behavior is performed or the time taken for the behavior, and outputs information regarding the selected solutions.

9. The information processing system of claim 2, wherein the output unit selects one or more solutions from a solution information storage unit that stores information regarding one or more solutions that can be proposed for at least one increase and decrease in at least one of the number of times the behavior or the amount of exercise, and outputs information regarding the selected solutions.

10. The information processing system of claim 8 or 9, further comprising a determination unit that determines the effectiveness or utilization of the solution by comparing the measurement data acquired in a first period before the proposal with the measurement data acquired in a second period after the proposal.

11. When the determination unit determines that the effect of the solution is low, The information processing system according to claim 10 , wherein the output unit outputs information about a solution different from the solution.

12. The information processing system according to claim 8 , wherein the output unit selects a predetermined number of the solutions from among one or more of the solutions that can be proposed in accordance with a priority order, and outputs information related to the selected solutions.

13. The information processing system according to claim 12 , wherein the output unit preferentially selects a solution that has a proven track record of effectiveness or a solution that has a proven track record of utilization in the past.

14. The processor: Obtaining measurement data relating to a pet to be detected; Detecting the number of times a specific behavior occurs and the time taken for the behavior based on the acquired measurement data, and detecting a health risk of the pet based on at least one of an increase and a decrease in the number of times the behavior occurs and at least one of an increase and a decrease in the time taken for the behavior; outputting information regarding the detected health risk of the pet; A method comprising:

15. The processor: Obtaining measurement data relating to a pet to be detected; Detecting the number of times a specific behavior occurs and the amount of movement based on the acquired measurement data, and detecting a health risk of the pet based on at least one of an increase and a decrease in the number of times the behavior occurs and at least one of an increase and a decrease in the amount of movement; outputting information regarding the detected health risk of the pet; A method comprising:

Citation Information

Patent Citations

  • Determination device, terminal device, determination system, program, and determination method

    JP2018174885A

  • Management server and management program for managing pet animal

    JP2020005534A

  • Health state determination device, health state determination system, and program

    JP2020005556A

  • Animal toilet, health condition determination device and user terminal

    JP6585523B2

  • JPP6630906B