JP7741211B2 - Pet Management System - Google Patents

Description

The present invention relates to a pet management system.

When dogs and cats suffer from dermatitis, they may exhibit head-shaking, scratching, and licking behaviors in response to itching. Owners of pet dogs and cats often take their pets to the veterinarian for treatment of their dermatitis, and receive medication. However, it is difficult for owners to determine the extent of their pet's itching, and it is also difficult to record all of their pet's behavior. As a result, owners are often unable to fully communicate their pet's symptoms to the veterinarian during their examination.

Various technologies for monitoring and managing pets are known. For example, Patent Document 1 describes monitoring a pet's indoor behavior in real time and identifying normal and abnormal behavior. Patent Document 2 describes providing a remotely monitored pet with food according to the pet's health condition. Furthermore, Patent Document 3 describes creating food determined based on the pet's obesity level information and the pet's preferences.

Patent No. 6626588 Patent No. 6807898 Patent No. 6538237

When a pet suffers from dermatitis, it would be desirable for the owner to be able to easily and quantitatively understand the behavior associated with the pet's itch. This would allow the owner to understand the feelings of their pet with dermatitis and live with their pet. Furthermore, when taking their pet to a veterinarian, if the owner could quantitatively understand the behavior associated with their pet's itch, they would be able to properly communicate their pet's symptoms to the veterinarian. This would allow the pet to be treated more appropriately.

The present invention was made in consideration of this background and aims to provide a pet management system that can easily and quantitatively grasp behaviors associated with pet itching.

One aspect of the present invention is a wearable device attached to a pet,
a management device configured to be able to communicate with the wearable device;
Equipped with
the wearable device includes a sensor that detects at least one of acceleration and angular velocity;
the wearable device or the management apparatus includes an itch behavior analysis unit that analyzes data detected by the sensor to generate itch behavior data representing behavior of the pet associated with itch;
The management device includes:
a management data storage unit for storing the itch behavior data;
a display unit that displays the itching behavior data stored in the management data storage unit, and the itching behavior analysis unit generates the itching behavior data using a machine learning model based on features selected from statistics of each frequency band divided into a plurality of frequency bands for the data detected by the sensor.

According to the present invention, owners can understand their pet's itching behavior by viewing the display of the management device. In particular, the itching behavior data displayed on the display is data obtained by analyzing at least one of the acceleration and angular velocity detected by the sensor. This allows owners to quantitatively understand their pet's itching behavior.

Furthermore, by having the wearable device attached to the pet at all times, the owner can understand the pet's itching behavior even when the owner is not watching the pet. Therefore, the owner can properly understand the pet's itching condition. As a result, the owner can live a life that is appropriate for the pet's itching symptoms. Furthermore, when taking the pet to the veterinarian, the owner can properly communicate the pet's symptoms to the veterinarian. As a result, the veterinarian can provide more appropriate treatment.

As described above, the above aspect makes it possible to provide a pet management system that can easily and quantitatively grasp behaviors associated with pet itching.

Note that the symbols in parentheses in the claims indicate their correspondence with the specific means described in the embodiments below, and do not limit the technical scope of the present invention.

FIG. 1 is a diagram showing a configuration of a pet management system according to a first embodiment. FIG. 2 is a diagram illustrating a functional configuration of a pet management system according to the first embodiment. 10 is a flowchart showing processing by an itch behavior analysis unit in the first embodiment. 10A to 10C are diagrams showing display screens for each operation of the owner terminal. FIG. 10 is a diagram illustrating the functional configuration of a pet management system according to a second embodiment. FIG. 11 is a diagram illustrating the functional configuration of a pet management system according to a third embodiment. 11 is a flowchart showing processing by an itch behavior analysis unit in embodiment 3. FIG. 10 is a diagram for explaining itching behavior data determination. 10 is a flowchart showing processing by an itch behavior analysis unit in embodiment 4. FIG. 10 is a diagram for explaining living activity data among candidate data. FIG. 13 is a diagram showing the configuration of a pet management system in a fifth embodiment. FIG. 13 is a diagram illustrating the functional configuration of a pet management system according to a fifth embodiment.

Pet management systems are used to easily and quantitatively understand behaviors associated with itching in pets such as dogs and cats. The pet management system includes at least a wearable device attached to the pet and a management device configured to be able to communicate with the wearable device.

The wearable device is equipped with a sensor that detects at least one of acceleration and angular velocity. The wearable device or the management device is equipped with an itch behavior analysis unit that analyzes data detected by the sensor to generate itch behavior data representing the pet's itch-related behavior. The management device also has a management data storage unit that stores the itch behavior data, and a display unit that displays the itch behavior data stored in the management data storage unit.

The itch behavior analysis unit may analyze the data detected by the sensor to distinguish between itch behavior data and daily behavior data that includes behavior associated with the pet walking or running, and generate itch behavior data that does not include daily behavior data. In this way, by distinguishing between itch behavior data and daily behavior data, the desired itch behavior data can be reliably generated.

The sensor may output itching behavior data and daily living behavior data as the same type of data, and the itching behavior analysis unit may analyze the data detected by the sensor to extract candidate data including the itching behavior data and daily living behavior data, distinguish between the itching behavior data and daily living behavior data based on the duration of the candidate data, and generate itching behavior data that does not include the daily living behavior data.

When itching behavior data and daily behavior data are output as the same type of data, the question arises as to how to distinguish between the two. While walking and running are continuous daily behaviors, itching behavior is instantaneous and less continuous than walking or running. Therefore, the itching behavior analysis unit focuses on the duration of candidate data that includes itching behavior data and daily behavior data, and distinguishes between itching behavior data and daily behavior data based on the duration of the candidate data. This allows for appropriate extraction of itching behavior data.

The itch behavior analysis unit may use a machine learning model to extract candidate data based on feature quantities selected from the statistics of each frequency band divided into multiple frequency bands for the data detected by the sensor. In this way, by using machine learning, it is possible to appropriately extract candidate data, which is the desired data, from a wide variety of data.

However, the input data (explanatory variable data) used to extract candidate data is important. Here, behaviors associated with itching include, for example, head shaking, scratching, and licking, which can be detected using acceleration and angular velocity. In such cases, appropriate candidate data can be extracted by dividing the frequency bands, using feature quantities selected from the statistics of each frequency band as explanatory variable data, and extracting candidate data using a machine learning model.

The itch behavior analysis unit may use a machine learning model to generate itch behavior data based on data detected by the sensor. The data detected by the sensor includes various data on behaviors other than those associated with itch. By using machine learning, itch behavior data can be appropriately extracted from the data detected by the sensor.

The itch behavior analysis unit may generate itch behavior data by comparing the data detected by the sensor with master data. This allows appropriate determination of whether or not the data detected by the sensor contains itch behavior data.

In this case, the master data may be a waveform pattern, and the itch behavior analysis unit may generate itch behavior data based on the similarity between the data detected by the sensor and the master data. This allows appropriate itch behavior data to be extracted.

The itch behavior data can also include the time the pet engaged in itch-related behavior or the number of times within a set period of time. The itch behavior analysis unit can analyze when the pet engaged in itch behavior by using data detected by the sensor. In this case, the owner can view the display unit to understand the time or number of times the pet engaged in itch-related behavior as itch behavior data. As a result, the owner can properly understand the pet's itch behavior.

The itching behavior data can also include the intensity level of the pet's itching behavior. When the itching is intense, the pet's itching behavior becomes more intense. Therefore, the itching behavior analysis unit can analyze the intensity level of the pet's itching behavior by using the data detected by the sensor. In this case, the owner can understand the intensity level of the itching behavior as itching behavior data by viewing the display unit. As a result, the owner can properly understand the pet's itching behavior.

The itch behavior analysis unit generates multiple types of itch behavior data, and the management device further includes an itch evaluation unit that performs itch evaluation processing based on the multiple types of itch behavior data to generate an itch index value that represents the degree of itch. The display unit can also display the itch behavior data stored in the management data storage unit and the itch index value generated by the itch evaluation unit. In other words, the itch index value is a value obtained by making a comprehensive judgment based on the multiple types of itch behavior data. In this case, the owner can understand the itch index value generated by the itch evaluation unit by viewing the display unit.

The itch evaluation unit can also generate an itch index value by performing itch evaluation processing based on the respective trends in multiple types of itch behavior data. Trends in itch behavior data refer to changes in itch behavior data over time. In other words, the itch index value generated by the itch evaluation unit takes into account trends in itch behavior data, i.e., changes in itch behavior data over time. This makes it possible to obtain an appropriate desired value as the itch index value.

The itch evaluation unit generates a degree of inflammation of dermatitis by performing itch evaluation processing based on the respective trends in multiple types of itch behavior data, and the display unit may display the itch behavior data stored in the management data storage unit, the itch index value generated by the itch evaluation unit, and the degree of inflammation. For example, when the degree of inflammation of dermatitis is in the early stages, licking behavior tends to increase, while scratching behavior tends to gradually increase as the inflammation progresses. Therefore, the degree of inflammation can be obtained based on the trends in licking behavior and scratching behavior. Then, by viewing the display unit, the owner can understand the degree of inflammation of their pet's dermatitis. The owner can understand whether the degree of inflammation of their pet is improving or worsening and can take appropriate measures.

The management device may also include a server configured to communicate with the wearable device, having the management data storage unit, and acquiring data stored in the wearable device, and an owner terminal configured to communicate with the server and having the display unit.

The management device may further include a bridge device configured to mediate communication between the wearable device and the server, and the server may obtain data stored in the wearable device through the bridge device.

The owner terminal may be configured to be able to communicate with the wearable device, and the server may be configured to be able to communicate with the owner terminal, and may be configured to obtain data stored in the wearable device through the owner terminal.

The owner terminal mediates data transmission between the wearable device and the server. In other words, the wearable device does not need to be equipped with high-performance communication equipment capable of communicating with the server; it is sufficient to have simple communication equipment capable of communicating with the owner terminal. This allows the wearable device to be low-cost and small-sized. Furthermore, data transmitted from the wearable device is stored on the server. Therefore, data can be centrally managed on the server, and people other than the owner can view the shared data.

The wearable device may be equipped with an itch behavior analysis unit, and the server may be equipped with a management data storage unit, and may acquire itch behavior data via the owner terminal.

In other words, the data transmitted by the wearable device is not data detected by the sensor, but itch behavior data generated by the itch behavior analysis unit. Compared to data detected by the sensor, itch behavior data has a significantly smaller data volume. Therefore, the amount of data transmitted by the wearable device can be reduced. Furthermore, the wearable device does not need to store the data detected by the sensor for a long period of time; it is sufficient to store the itch behavior data. This allows the wearable device to reduce its data storage capacity.

The server may be equipped with a management data storage unit and an itching behavior analysis unit, and may acquire data detected by the sensor via the owner terminal.

In this case, the wearable device does not need to be equipped with an itch behavior analysis unit. Therefore, the wearable device does not require a high-performance processing device, making it low-cost. The server is capable of performing high-performance computational processing. Therefore, by having the server analyze itch behavior, high-performance computational processing, such as computational processing using a machine learning model, can be performed.

The owner terminal may further include a behavioral information input unit configured to input pet behavioral information data in association with the itching behavior data displayed on the display unit, and the server's management data storage unit may store the behavioral information data input by the behavioral information input unit in association with the itching behavior data. In other words, the owner can input pet behavioral information data based on their own senses using the behavioral information input unit. The server's management data storage unit stores the itching behavior data generated from the sensor detection data and the behavioral information data input by the owner. Therefore, by viewing the display unit of the owner terminal, the owner can grasp both the quantitative itching behavior data and the behavioral information data input by the owner.

Furthermore, a veterinarian terminal may be provided that is configured to be able to communicate with the server and to be able to view the itching behavior data stored in the management data storage unit. By viewing the veterinarian terminal, the veterinarian can easily understand the pet's condition. Furthermore, the itching behavior data viewed by the veterinarian is the same as the itching behavior data that can be viewed by the owner. Therefore, information can be shared between the owner and the veterinarian, and appropriate communication can be achieved between the two parties.

The owner terminal may include a medical interview input unit configured to input medical interview data for a veterinarian's visit, the server's management data storage unit may store the medical interview data input by the medical interview input unit, and the veterinarian terminal may be configured to allow the medical interview data to be viewed. By inputting the medical interview data into the owner terminal, the owner can share the medical interview data with the veterinarian when the pet visits the veterinarian.

Itching behaviors should include at least one of head shaking, scratching, and licking. These behaviors can be used to identify itching behaviors.

(Embodiment 1)
A pet management system 1 according to a first embodiment will be described below with reference to the drawings.

1. Overview of the Configuration of the Pet Management System 1 The configuration of the pet management system 1 will be described with reference to Fig. 1. As shown in Fig. 1, the pet management system 1 is configured to include a wearable device 3 attached to a pet 2, a management device 4, and a veterinarian terminal 5a in a veterinary clinic 5.

The pet 2 to be managed is, for example, a companion animal such as a dog or cat. When pet 2 suffers from dermatitis, it exhibits behaviors associated with itching. Examples of behaviors pet 2 exhibited by itching include head shaking, scratching, and licking. Furthermore, when the level of itching is severe, pet 2's behaviors associated with itching tend to increase in frequency and intensity. For example, when pet 2 performs head shaking, the more severe the itching, the more frequently and vigorously it shakes its head. The same is true for scratching and licking.

It is also known that Pet 2's behavior associated with itching changes depending on the progression of the dermatitis inflammation. For example, when the dermatitis inflammation is in the early stages, there is a tendency for licking behavior to increase, and as the inflammation progresses, there is a tendency for scratching behavior to gradually increase.

The wearable device 3 is a device that can be attached to the pet 2. The wearable device 3 is a device that is attached to a pet wearing device such as a collar, a pet harness, or pet wear. The wearable device 3 may be built into the pet wearing device or may be a separate component attached to the pet wearing device. In Figure 1, the wearable device 3 is illustrated as a separate component attached to the collar.

Wearable device 3 has a built-in sensor and can detect physical quantities that occur in conjunction with the behavior of pet 2. Furthermore, wearable device 3 can perform simple analysis processing and communicate with external devices. For example, as a simple analysis process, wearable device 3 uses data detected by the sensor to generate itch behavior data that represents the behavior of the pet associated with itching. Wearable device 3 can also perform short-range wireless communication using Bluetooth (registered trademark) and NFC (registered trademark).

The management device 4 is configured to be able to communicate with the wearable device 3. The management device 4 comprises an owner terminal 4a and a server 4b. The owner terminal 4a is owned by the owner and is configured to be able to communicate with the wearable device 3. When the owner terminal 4a is brought close to the wearable device 3, it can perform short-range wireless communication, for example, using Bluetooth (registered trademark) or NFC (registered trademark). However, the wearable device 3 may also be configured to be able to connect to the Internet. Furthermore, the owner terminal 4a and the server 4b form a network via the Internet. Therefore, the owner terminal 4a and the server 4b are configured to be able to communicate via the Internet. The server 4b is configured to be able to communicate with the wearable device 3 via the owner terminal 4a.

The server 4b stores the data transmitted from the wearable device 3 via the owner terminal 4a. Furthermore, the server 4b can also perform itch evaluation processing based on the acquired data. The server 4b also displays the display content on the owner terminal 4a by performing processing to display display content that can be viewed on the owner terminal 4a. This display content includes itch behavior data, itch index values, etc.

The veterinarian terminal 5a is installed in the animal hospital 5 or is portable and operated by the veterinarian. The veterinarian terminal 5a and server 4b form a network via the Internet, and therefore the veterinarian terminal 5a and server 4b are configured to be able to communicate via the Internet. The veterinarian terminal 5a can display content processed by the server 4b. The content displayed on the veterinarian terminal 5a and the content displayed on the owner terminal 4a may be the same or different. Even if the display content is different, only the display format is different, and the information source is common data stored on the server 4b.

2. Detailed Configuration of Pet Management System 1 The detailed configuration of the pet management system 1 will be described with reference to Fig. 2. In particular, the wearable device 3, owner terminal 4a, server 4b, and veterinarian terminal 5a that make up the pet management system 1 will be described in detail.

The wearable device 3 is configured with an acceleration sensor 11, an angular velocity sensor 12, a data storage unit 13, an itch behavior analysis unit 14, and a communication processing unit 15. The acceleration sensor 11 is a sensor capable of detecting acceleration on three orthogonal axes. The angular velocity sensor 12 is a sensor capable of detecting angular velocity in rotation around each of the three orthogonal axes. While this embodiment exemplifies a configuration in which the acceleration sensor 11 and the angular velocity sensor 12 are included, the device may also be configured with the acceleration sensor 11 but not the angular velocity sensor 12, or with the angular velocity sensor 12 but not the acceleration sensor 11. These configurations can be appropriately determined depending on the detection accuracy and the detection purpose. Furthermore, for the purpose of higher detection accuracy, the wearable device 3 may also be configured with additional sensors other than the acceleration sensor 11 and the angular velocity sensor 12, such as a geomagnetic sensor.

The data storage unit 13 stores detection data 13a, itch behavior data 13b, and master data 13c. The detection data 13a is data detected by the acceleration sensor 11 and data detected by the angular velocity sensor 12. The detection data 13a is acceleration data at each detection time and angular velocity data at each detection time. Here, the detection data 13a stored in the data storage unit 13 can be erased after being used for analysis processing by the itch behavior analysis unit 14, which will be described later. For example, the data storage unit 13 is configured with a ring buffer of a predetermined capacity for storing the detection data 13a, and when the predetermined capacity is reached, new detection data 13a is overwritten.

Itching behavior data 13b is data generated by the itch behavior analysis unit 14 through analysis using the detection data 13a. Itching behavior data 13b is data representing behaviors associated with itching by pet 2. Itching behavior data 13b includes at least one of head-shaking behavior, scratching behavior, and licking behavior as behaviors associated with itching. In this embodiment, itch behavior data 13b includes all three elements: head-shaking behavior, scratching behavior, and licking behavior. More specifically, itch behavior data 13b includes, for example, itch behavior data representing itch behavior for a predetermined period of time (e.g., 1 to 30 seconds), itch behavior data representing itch behavior for each hour, and itch behavior data representing itch behavior for each day. Furthermore, the data capacity of itch behavior data 13b is smaller than the data capacity of detection data 13a when compared over the same period of time. For example, the data storage unit 13 is configured by a ring buffer or the like with a predetermined capacity for storing the itch behavior data 13b, and when the predetermined capacity is reached, new itch behavior data 13b is overwritten.

Master data 13c is data used in the analysis process by the itch behavior analysis unit 14. Master data 13c is a waveform pattern for the acceleration sensor 11 and angular velocity sensor 12 when pet 2 performs an itch-related behavior. In this embodiment, master data 13c is data for a predetermined time (e.g., 1 to 30 seconds) equal to the smallest unit time of itch behavior data 13b.

For example, one of the master data 13c is a waveform pattern representing acceleration over time and a waveform pattern representing angular velocity over time for head-shaking behavior. The master data 13c representing head-shaking behavior is set to different waveform patterns depending on the direction of the head-shaking, the strength of the head-shaking, etc. In other words, the data storage unit 13 stores multiple master data 13c for head-shaking behavior.

Furthermore, for scratching and licking behaviors, as with head-shaking behaviors, master data 13c is set using waveform patterns representing acceleration over time and waveform patterns representing angular velocity over time. Similarly for scratching and licking behaviors, master data 13c is set using different waveform patterns depending on the area and strength of the scratching behavior and the area and strength of the licking behavior.

The waveform pattern representing master data 13c is set by the amplitude and frequency of acceleration, the amplitude and frequency of angular velocity, etc. These elements indicate different ranges for each of head-shaking behavior, scratching behavior, and licking behavior, and also indicate different ranges depending on the strength of each behavior.

The itch behavior analysis unit 14 analyzes the detection data 13a stored in the data storage unit 13 to generate itch behavior data 13b representing behaviors associated with itch of the pet 2. The itch behavior analysis unit 14 analyzes the detection data 13a at set intervals (e.g., 1 to 30 seconds) to generate itch behavior data 13b.

Specifically, the itch behavior analysis unit 14 compares the detection data 13a for a predetermined time (e.g., 1 to 30 seconds) with multiple master data 13c stored in the data storage unit 13 to determine whether the detection data 13a for that predetermined time corresponds to one of multiple itch-related behaviors or to no itch-related behavior. The comparison between the detection data 13a and the master data 13c performed by the itch behavior analysis unit 14 can, for example, determine the similarity of changes in data values, and determine that the two match if the similarity is equal to or greater than a predetermined value. The itch behavior analysis unit 14 can also compare both data as images, and determine that the two match if the image similarity is equal to or greater than a predetermined value using machine learning or the like.

For example, the itch behavior analysis unit 14 can determine the type of behavior associated with itch by comparing the amplitude and frequency of acceleration, the amplitude and frequency of angular velocity, etc. over a set predetermined period of time (e.g., 1 to 30 seconds), and can further determine the intensity level of the behavior.

Here, let us assume that the itching behavior analysis unit 14 performs an analysis using the detection data 13a for a predetermined time (e.g., 1 to 30 seconds) and finds that the pet 2 is engaged in scratching behavior during that predetermined time, and that the maximum intensity level of the scratching behavior is "5." In this case, the itching behavior analysis unit 14 determines that the pet 2 is engaging in scratching behavior at an intensity level of "5" for the entire period of that predetermined time.

Furthermore, by continuously accumulating itch behavior data for a predetermined period of time (e.g., 1 to 30 seconds), the itch behavior analysis unit 14 can generate the time and number of times each of multiple types of itch-related behaviors was performed, for example, within one hour or one day. The itch behavior analysis unit 14 is not limited to setting the set period to one hour or one day, but can set it to any time.

Furthermore, the itch behavior analysis unit 14 uses the itch behavior data 13b for each predetermined time period (e.g., 1 to 30 seconds) to generate hourly itch behavior data and daily itch behavior data. For example, the hourly itch behavior data may be data summarizing a predetermined time period (e.g., 1 to 30 seconds) for one hour, or may be data having one value for one hour. In the latter case, the hourly itch behavior data uses the most frequent behavior during that hour as the target behavior for that period, and the maximum intensity level of that behavior as the intensity level for that period. Then, after generating the itch behavior data 13b, the itch behavior analysis unit 14 stores it in the data storage unit 13.

The communication processing unit 15 is configured to communicate with an external device, specifically the owner terminal 4a. The communication processing unit 15 performs short-range wireless communication such as Bluetooth (registered trademark) or NFC (registered trademark). When a data transmission request is made from the owner terminal 4a, the communication processing unit 15 transmits the itch behavior data 13b stored in the data storage unit 13 to the owner terminal 4a.

The owner terminal 4a is configured to include a communication processing unit 21, a data acquisition unit 22, a display unit 23, a behavioral information input unit 24, and a medical interview input unit 25. The communication processing unit 21 is configured to be able to communicate with the communication processing unit 15 of the wearable device 3. In particular, the communication processing unit 21 performs short-range wireless communication such as Bluetooth (registered trademark) or NFC (registered trademark). Furthermore, the communication processing unit 21 is configured to be able to communicate with the communication processing unit 31 of the server 4b (described later) via the Internet.

Operated by the owner, the data acquisition unit 22 performs a process to acquire the itch behavior data 13b stored in the data storage unit 13 of the wearable device 3. However, the owner terminal 4a does not store the data, but acts as an intermediary to transfer the itch behavior data 13b received from the wearable device 3 to the server 4b.

The display unit 23 displays various display contents in response to the owner's operation. The display unit 23 displays itch behavior data, itch index values, inflammation progression levels, etc. stored on the server 4b. The display contents of the display unit 23 will be described later.

The behavioral information input unit 24 is configured to input behavioral information data of the pet 2 in association with the itching behavior data 13b displayed on the display unit 23, etc. The owner can freely input text, photos, illustrations, etc. about what the owner saw and felt as the behavioral information data to be input. The behavioral information data input by the behavioral information input unit 24 is stored on the server 4b via the communication processing unit 21. The medical interview input unit 25 is configured to input medical interview data for a veterinarian's visit. The medical interview data input by the medical interview input unit 25 is stored on the server 4b via the communication processing unit 21.

Server 4b is, for example, a cloud server. Server 4b is configured to include a communication processing unit 31, a management data storage unit 32, an itch evaluation unit 33, and a display processing unit 34. The communication processing unit 31 is configured to be able to communicate with the owner terminal 4a and the veterinarian terminal 5a via the Internet.

The management data storage unit 32 stores itch behavior data 32a, itch index value 32b, inflammation progression level 32c, behavioral information data 32d, and medical interview data 32e.

Itching behavior data 32a is equal to itch behavior data 13b stored in the data storage unit 13 of the wearable device 3. When the owner operates the data acquisition unit 22 of the owner terminal 4a, the itch behavior data 13b stored in the data storage unit 13 of the wearable device 3 is stored as itch behavior data 32a in the management data storage unit 32 of the server 4b. In other words, the server 4b, via the owner terminal 4a, acquires the itch behavior data 13b stored in the data storage unit 13 of the wearable device 3 and stores it as itch behavior data 32a in the management data storage unit 32.

The itch index value 32b is a value that represents the degree of itch of the pet 2. The itch index value 32b is expressed, for example, as "Level 1" to "Level 5." The itch index value is generated by the itch evaluation unit 33 through an itch evaluation process based on multiple types of itch behavior data.

Inflammation progression level 32c indicates the degree of progression of dermatitis when pet 2 is suffering from dermatitis. Inflammation progression level 32c is generated by the itch evaluation unit 33 performing an itch evaluation process based on multiple types of itch behavior data.

The behavioral information data 32d is data input by the behavioral information input unit 24 of the owner terminal 4a. The behavioral information data 32d is stored in association with the itching behavior data displayed on the display unit 23 of the owner terminal 4a, as input by the behavioral information input unit 24 of the owner terminal 4a. The medical interview data 32e is data input by the medical interview input unit 25 of the owner terminal 4a.

The itch evaluation unit 33 performs itch evaluation processing based on multiple types of itch behavior data 32a stored in the management data storage unit 32. In particular, the itch evaluation unit 33 performs itch evaluation processing based on the trends in each of the multiple types of itch behavior data 32a. The trends in the itch behavior data 32a refer to changes in the itch behavior data 32a over time.

The itch evaluation unit 33 performs an itch evaluation process to generate an itch index value 32b that represents the degree of itch. Specifically, the itch evaluation unit 33 generates the itch index value 32b by making a comprehensive judgment based on the respective trends in multiple types of itch behavior data 32a. For example, the itch evaluation unit 33 generates the itch index value 32b by applying machine learning, statistical methods, etc. In other words, the itch index value 32b generated by the itch evaluation unit 33 is a value that takes into account the trends in the itch behavior data, i.e., changes over time in the itch behavior data 32a. The itch evaluation unit 33 stores the generated itch index value 32b in the management data storage unit 32. This makes it possible to obtain an appropriate desired value as the itch index value.

Furthermore, the itch evaluation unit 33 generates the dermatitis inflammation progression level 32c by performing an itch evaluation process. Specifically, the itch evaluation unit 33 generates the inflammation progression level 32c by comprehensively assessing the progress of each of the multiple types of itch behavior data 32a. For example, the itch evaluation unit 33 generates the inflammation progression level 32c by applying machine learning, statistical methods, etc. In other words, the itch evaluation unit 33 can capture changes in the behavior of the pet 2 based on the progress of each of the multiple types of itch behavior data 32a, and can obtain the dermatitis inflammation progression level 32c based on the changes in the behavior of the pet 2. For example, when the inflammation progression level of dermatitis is in the early stages, licking behavior tends to increase, whereas scratching behavior tends to gradually increase as the inflammation progresses. Therefore, the inflammation progression level can be obtained based on the progress of licking behavior, scratching behavior, etc. The itch evaluation unit 33 stores the generated inflammation progression level 32c in the management data storage unit 32.

The display processing unit 34 executes display processing in response to display requests from the owner terminal 4a and the veterinarian terminal 5a. The display processing unit 34 generates display data in response to the display request using the itch behavior data 32a, itch index value 32b, inflammation progression level 32c, behavior information data 32d, and medical interview data 32e stored in the management data storage unit 32. The display processing unit 34 transmits the display data to the owner terminal 4a and the veterinarian terminal 5a via the communication processing unit 31.

The veterinarian terminal 5a is configured to include a communications processing unit 41, an input unit 42, and a display unit 43. The communications processing unit 41 is configured to be able to communicate with the server 4b via the Internet. The input unit 42 allows the veterinarian to input various display requests. The display unit 43 receives and displays display data corresponding to the input from the input unit 42 from the server 4b. Therefore, the display unit 43 of the veterinarian terminal 5a is configured to allow viewing of the pet 2's itch behavior data 32a, itch index value 32b, inflammation progression level 32c, behavior information data 32d, and medical interview data 32e. The display content displayed on the display unit 43 may be the same as the display content displayed on the owner terminal 4a, or may be in a different format.

3. Example of Processing by the Itch Behavior Analysis Unit 14 As described above, the itch behavior analysis unit 14 constitutes part of the wearable device 3, and generates itch behavior data 13b based on the detection data 13a. An example of processing by the itch behavior analysis unit 14 will be described with reference to FIG. 3.

The itch behavior analysis unit 14 first acquires detection data 13a for a predetermined period of time (e.g., 1 to 30 seconds, same below) from the data storage unit 13 (S1). Next, the itch behavior analysis unit 14 compares the acquired detection data 13a with the master data 13c stored in the data storage unit 13 to identify the itch behavior (S2). Specifically, the itch behavior analysis unit 14 selects the type of itch behavior from, for example, head-shaking behavior, scratching behavior, licking behavior, etc. Furthermore, the itch behavior analysis unit 14 identifies the intensity level of the corresponding behavior.

Next, the itch behavior analysis unit 14 stores the itch behavior data 13b in the data storage unit 13 (S3). The itch behavior data 13b stored at this time includes the type of itch behavior, the start and end times of the itch behavior, the intensity level of the itch behavior, etc.

The itch behavior analysis unit 14 then determines whether the hour has been reached (S4). For example, 1:00, 2:00, etc. If the hour has been reached (S4: Yes), the itch behavior analysis unit 14 determines the intensity level for the most recent hourly unit and stores this as itch behavior data 13b in the data storage unit 13 (S5). If the hour has not been reached (S4: No), the process proceeds to the next step without performing S5.

Next, the itch behavior analysis unit 14 determines whether 24:00 has been reached (S6). If 24:00 has been reached (S6: Yes), the itch behavior analysis unit 14 determines the intensity level for the most recent day and stores it as itch behavior data 13b in the data storage unit 13 (S7). If 24:00 has not been reached (S6: No), the process proceeds to the next step without performing S7.

Then, the process returns to S1 and is repeated. In other words, until one hour has passed, only itch behavior data 13b is generated for each specified hour. Then, with each passing hour, itch behavior data 13b for each hour is added. Furthermore, with each passing day, itch behavior data 13b for each day is added.

4. Display Screen of Owner Terminal 4a The display screen of the owner terminal 4a will be described with reference to Fig. 4. The owner terminal 4a is, for example, a touch panel type mobile terminal such as a smartphone or tablet terminal. The display screen of the owner terminal 4a functions as the display unit 23 shown in Fig. 2, as well as the data acquisition unit 22, the behavioral information input unit 24, and the medical interview input unit 25.

The "Home" icon 61, "View Data" icon 62, and "Settings" icon 63 are displayed in the lower section of the home screen 51 of the owner terminal 4a. The upper area 71 of the home screen 51 displays a photo and name of the pet 2. The central area of the home screen 51 displays a "Load Data" button 72, a "Enter Behavioral Information" button 73, and a "Enter Medical Questionnaire" button 74.

When the owner taps the "View Data" icon 62, the data viewing screen 52 is displayed. The data viewing screen 52 displays a score as "Today's Itch Index Value." The itch index value is the itch index value 32b stored in the management data storage unit 32 of the server 4b. The itch index value 32b is a value obtained by making a comprehensive judgment based on multiple types of itch behavior data. The owner can understand the itch index value by viewing the score for "Today's Itch Index Value."

Furthermore, the data viewing screen 52 displays the time (hours per day) and intensity level for each of several types of itch behaviors as "Today's Itch Behavior." The displayed itch behaviors include head-shaking behavior, scratching behavior, and licking behavior. By viewing the time or number of times in "Today's Itch Behavior," the owner can understand the time or number of times that itch-related behaviors were performed. As a result, the owner can properly understand Pet 2's itch behavior. Furthermore, by viewing the intensity level section in "Today's Itch Behavior," the owner can understand the intensity level of itch-related behavior. As a result, the owner can properly understand Pet 2's itch behavior.

Furthermore, the data viewing screen 52 displays a score as the "inflammation progression level," as well as arrows indicating improvement, no change, or worsening. The inflammation progression level is the inflammation progression level 32c stored in the management data storage unit 32 of the server 4b. An upward diagonal arrow indicates worsening, a horizontal arrow indicates no change, and a downward diagonal arrow indicates improvement. By viewing the "inflammation progression level," the owner can understand the inflammation progression level of pet 2's dermatitis. Furthermore, the owner can understand whether pet 2's inflammation level is improving or worsening, and can take appropriate measures.

When the owner taps on one of the "Today's Itching Behavior" options on the data viewing screen 52, the data viewing details screen 53 is displayed. The data viewing details screen 53 displays a time series graph of the itching behavior. The displayed graph can be selected by "amount (time)" and "intensity level," and the time period can be selected from "today," "1 week," and "1 month." The data viewing details screen 53 in Figure 4 displays a graph of the time of today's head-shaking behavior. The same is true for scratching behavior and licking behavior. In addition, by scrolling downwards on the data viewing details screen 53, numerical values corresponding to the graph are displayed.

In addition, when the owner taps the "Read Data" button 72 on the home screen 51 while holding the owner terminal 4a close to the wearable device 3, the itch behavior data 13b stored in the data storage unit 13 of the wearable device 3 is uploaded to the management data storage unit 32 of the server 4b via the owner terminal 4a.

In this way, the object of communication between the wearable device 3 and the owner terminal 4a, and between the owner terminal 4a and the server 4b, is the itch behavior data stored in the data storage unit 13 of the wearable device 3. In other words, the itch behavior data that is the object of communication has a smaller data volume than the detection data 13a detected by the acceleration sensor 11 and the angular velocity sensor 12. Therefore, the communication volume can be reduced.

Furthermore, when the owner taps the "Enter Behavioral Information" button 73 on the home screen 51, the behavioral information input screen 54 is displayed. The owner can freely input text, photos, illustrations, etc. on the behavioral information input screen 54. In this way, the owner can input behavioral information data of the pet 2 based on their own perceptions. Furthermore, although not shown, the owner can view behavioral information data that has been input in the past. Therefore, the owner can understand the quantitative itch behavior data together with the behavioral information data that the owner has input.

In addition, when the owner taps the "Enter medical interview" button 74 on the home screen 51, the medical interview input screen 55 is displayed. The medical interview input screen 55 allows the owner to enter information necessary for the veterinarian's examination. By entering the medical interview data on the owner terminal, the owner can share the medical interview data with the veterinarian when the pet is examined by the veterinarian.

5. Effects of the embodiment According to the pet management system 1 of this embodiment, the owner can understand the itching behavior of the pet 2 by viewing the display unit 23 of the owner terminal 4a of the management device 4. In particular, the itching behavior data displayed on the display unit 23 is data obtained by analyzing data detected by the acceleration sensor 11 and the angular velocity sensor 12. Therefore, the owner can quantitatively understand the itching behavior of the pet 2.

Furthermore, by having the wearable device 3 attached to the pet 2 at all times, the owner can understand the itching behavior of the pet 2 even when the owner is not watching the pet 2. Therefore, the owner can properly understand the itching condition of the pet 2. As a result, the owner can live a life that is appropriate to the itching symptoms of the pet 2. Furthermore, when the pet 2 is examined by a veterinarian, the owner can properly communicate the symptoms of the pet 2 to the veterinarian. As a result, the veterinarian can provide more appropriate treatment. Therefore, the behavior associated with the itching of the pet 2 can be easily and quantitatively understood.

Furthermore, in the pet management system 1, the veterinarian terminal 5a is configured to be able to communicate with the server 4b and to be able to view the itching behavior data 32a stored in the management data storage unit 32. By viewing the veterinarian terminal 5a, the veterinarian can easily understand the condition of the pet 2. Furthermore, the itching behavior data 32a viewed by the veterinarian is the same as the itching behavior data 32a that can be viewed by the owner. Therefore, information can be shared between the owner and the veterinarian, and appropriate communication can be achieved between the two parties.

(Embodiment 2)
The pet management system 100 of the second embodiment will be described with reference to Fig. 5. Note that, among the reference numerals used in the second and subsequent embodiments, the same reference numerals as those used in the previous embodiments represent the same components as those in the previous embodiments unless otherwise specified.

In embodiment 1, the wearable device 3 includes an itch behavior analysis unit 14, and the data storage unit 13 of the wearable device 3 stores itch behavior data 13b and master data 13c. In this embodiment, as shown in FIG. 5, the server 4b constituting the management device 4 includes the itch behavior analysis unit 14. In this case, the management data storage unit 32 of the server 4b constituting the management device 4 stores the detection data 13a, itch behavior data 13b, and master data 13c. The data storage unit 13 of the wearable device 3 stores the detection data 13a, but does not store the itch behavior data or master data.

In this case, the wearable device 3 does not need to be equipped with the itch behavior analysis unit 14. Therefore, the wearable device 3 does not require a high-performance processing device, making it low-cost. The server 4b is capable of performing high-performance arithmetic processing. Therefore, by having the server 4b analyze the itch behavior, high-performance arithmetic processing can be performed.

In addition, the owner terminal 4a constituting the management device 4 may be equipped with an itch behavior analysis unit 14. In this case, the owner terminal 4a constituting the management device 4 will store the detection data 13a, itch behavior data 13b, and master data 13c.

(Embodiment 3)
A pet management system 200 according to a third embodiment will be described with reference to Fig. 6 to Fig. 8. As shown in Fig. 6, in the pet management system 200 according to this embodiment, detection data 13a is stored in the data storage unit 13 of the wearable device 3, similar to the second embodiment.

The server 4b acquires the detection data 13a via the owner terminal 4a. Therefore, the management data storage unit 232 of the server 4b stores the acquired detection data 232f. Note that the detection data 232f is the same as the detection data 13a detected by the sensors 11 and 12. The management data storage unit 232 further stores a machine learning model 232g for use in analyzing itch behavior.

The following describes machine learning model 232g. Machine learning model 232g is a model that outputs objective variable data when explanatory variable data is input. For example, in machine learning model 232g, the explanatory variable data is a feature generated from detection data 232f, and the objective variable data is a judgment score that represents behavior associated with itching.

Feature quantities as explanatory variable data are, for example, feature quantities selected from the statistics of each frequency band divided into multiple frequency bands for the detection data 232f. The multiple frequency bands can be divided arbitrarily, for example, dividing the target frequency band into intervals of 0.5 Hz to 3 Hz. The statistical quantity can be at least one selected from the minimum value, maximum value, median, first quartile, second quartile, mean value (arithmetic mean, geometric mean, harmonic mean), difference between maximum and minimum values, variance, standard deviation, skewness, kurtosis, etc. However, it is preferable to use multiple statistical quantities.

The judgment score as the objective variable data indicates that, for example, a smaller value indicates that the behavior is not associated with itching, and a larger value indicates that the behavior is associated with itching. The judgment score can be set within any range of values.

The machine learning model 232g is obtained by learning using previously prepared training data and applying a known machine learning method. Any machine learning method can be applied, such as supervised learning, unsupervised learning, or reinforcement learning. For example, a one-class support vector machine can be applied as a type of unsupervised learning method as a machine learning method. However, other methods may also be applied.

Furthermore, as mentioned above, types of itching behavior include, for example, head-shaking behavior, scratching behavior, and licking behavior. Therefore, it is preferable that the machine learning model 232g be a different model for each type of itching behavior. For example, a model for head-shaking behavior, a model for scratching behavior, a model for licking behavior, etc. are stored. In this case, it is preferable that the feature values also be different for each type of itching behavior.

The server 4b also includes an itch behavior analysis unit 214. The itch behavior analysis unit 214 generates itch behavior data 32a representing behaviors associated with itch of the pet 2 by analyzing the detection data 232f stored in the management data storage unit 232. The itch behavior analysis unit 214 extracts the itch behavior data 32a based on the detection data 232f using a machine learning model.

As shown in FIG. 7, the itch behavior analysis unit 214 first acquires detection data 232f for a predetermined period of time (e.g., 1 to 30 seconds, same below) from the management data storage unit 232 (S11). Next, the itch behavior analysis unit 214 generates feature quantities of the detection data 232f as explanatory variable data to be used in the inference phase of machine learning (S12). If there are multiple types of itch behavior, the itch behavior analysis unit 214 generates feature quantities according to the type of itch behavior.

Next, the itch behavior analysis unit 214 uses the machine learning model 232g based on the generated features to determine whether the data is itch behavior data 32a (S13).

In detail, by inputting features into the machine learning model 232g, a judgment score is output. For example, as shown in Figure 8, a judgment score is output at each time. If the output judgment score exceeds the threshold value Th, it is determined to be itch behavior data 32a, and if it is equal to or less than the threshold value Th, it is determined not to be itch behavior data 32a. In Figure 8, there are three judgment scores that exceed the threshold value Th. These are determined to be behavior data associated with itch.

Furthermore, the itch behavior analysis unit 214 uses the judgment score to identify the intensity level of the itch behavior. If there are multiple types of itch behavior, the itch behavior analysis unit 214 makes a judgment for each type of itch behavior.

Next, the itch behavior analysis unit 214 stores the itch behavior data 32a in the data storage unit 13 (S14). The itch behavior data 32a stored at this time includes the type of itch behavior, the start and end times of the itch behavior, the intensity level of the itch behavior, etc.

Next, the itch behavior analysis unit 214 determines whether the hour has been reached (S15). For example, 1:00, 2:00, etc. If the hour has been reached (S15: Yes), the itch behavior analysis unit 214 determines the intensity level for the most recent hourly unit and stores it as itch behavior data 32a in the management data storage unit 232 (S16). If the hour has not been reached (S15: No), the unit proceeds to the next step without performing S16.

Next, the itch behavior analysis unit 214 determines whether 24:00 has been reached (S17). If 24:00 has been reached (S17: Yes), the itch behavior analysis unit 214 determines the intensity level for the most recent day and stores it as itch behavior data 32a in the management data storage unit 232 (S18). If 24:00 has not been reached (S17: No), the process proceeds to the next step without performing the process of S18.

Then, the process returns to S11 and is repeated. In other words, until one hour has passed, only itching behavior data 32a is generated for each predetermined time period. Then, with each passing hour, hourly itching behavior data 32a is added. Furthermore, with each passing day, daily itching behavior data 32a is added.

As described above, the itch behavior analysis unit 214 generates itch behavior data 32a using a machine learning model based on the detection data 232f. The detection data 232f includes various data on behaviors other than those associated with itch. By using machine learning, it is possible to appropriately extract itch behavior data 32a from the detection data 232f.

(Embodiment 4)
A pet management system 200 according to the fourth embodiment will be described. The functional configuration of the pet management system 200 according to the fourth embodiment is the same as that according to the third embodiment. However, the processing of the itch behavior analysis unit 214 is different.

In this embodiment, the itch behavior analysis unit 214 analyzes the detection data 232f to distinguish between itch behavior data 32a and living behavior data including behavior associated with walking or running of the pet 2, and generates itch behavior data 32a that does not include living behavior data.

Here, sensors 11 and 12 output itch behavior data 32a and daily living behavior data as the same type of data. In other words, the detection data 232f detected by sensors 11 and 12 is a data type that cannot be distinguished between itch behavior data 32a and daily living behavior data.

Therefore, the itch behavior analysis unit 214 first analyzes the detection data 232f to extract candidate data that includes itch behavior data and daily living behavior data. In other words, since it is not possible to distinguish between itch behavior data and daily living behavior data at this stage, the candidate data will be data that includes itch behavior data and daily living behavior data.

The itch behavior analysis unit 214 applies machine learning to extract candidate data. That is, the itch behavior analysis unit 214 extracts candidate data using a machine learning model 232g based on feature quantities selected from statistics for each of multiple frequency bands divided into multiple frequency bands for the detection data 232f. The machine learning model 232g is substantially similar to the machine learning model 232g of embodiment 3.

Furthermore, the itch behavior analysis unit 214 distinguishes between itch behavior data 32a and daily living behavior data from the candidate data, and generates itch behavior data 32a that does not include daily living behavior data. At this time, the itch behavior analysis unit 214 takes into account the duration of the candidate data. In other words, the itch behavior analysis unit 214 distinguishes between itch behavior data and daily living behavior data based on the duration of the candidate data.

Daily living activities such as walking and running are not instantaneous activities, but are activities that continue for a certain period of time. On the other hand, itch behavior is an activity that occurs over a shorter period of time than walking or running. Therefore, the itch behavior analysis unit 214 determines that candidate data is daily living activity data if the duration of the data is long, and determines that it is itch behavior data if the duration is short.

The processing of the itch behavior analysis unit 214 will be explained with reference to Figures 8 to 10. As shown in Figure 9, the itch behavior analysis unit 214 first acquires detection data 232f for a predetermined period of time (e.g., 1 to 30 seconds, same below) from the management data storage unit 232 (S21). Next, the itch behavior analysis unit 214 generates feature quantities of the detection data 232f as explanatory variable data to be used in the inference phase of machine learning (S22). If there are multiple types of itch behavior, the itch behavior analysis unit 214 generates feature quantities according to the type of itch behavior.

Next, the itch behavior analysis unit 214 uses the machine learning model 232g to determine whether the data is candidate data (including itch behavior data and daily living behavior data) based on the generated features (S23).

In detail, by inputting features into the machine learning model 232g, a judgment score is output. For example, when pet 2 is performing an action associated with itching, a judgment score is output at each time, as shown in Figure 8. Furthermore, when pet 2 is walking or running, a judgment score is output at each time, as shown in Figure 10. As shown in Figures 8 and 10, the judgment score exceeds the threshold value Th in both cases of an action associated with itching and walking or running.

Next, the itch behavior analysis unit 214 distinguishes between itch behavior data 32a and daily living behavior data based on the duration of the candidate data and extracts the itch behavior data 32a excluding the daily living behavior data (S24).

8 and 10, it can be seen that the itch behavior data 32a has a short duration, while the daily life behavior data has a long duration. Therefore, if the duration of the candidate data exceeds a predetermined duration threshold, it is determined to be itch behavior data 32a.

However, people may take temporary breaks during walks, etc. If the length of the break is less than a predetermined stop time threshold, the break may be considered to be part of a walk. In this case, the behavior during the break will be included in daily activities such as walking and running. In other words, even if the judgment score during the break indicates a value that exceeds the threshold Th, it will not be determined to be itch behavior data 32a.

Furthermore, the itch behavior analysis unit 214 identifies the intensity level of the itch behavior using the judgment score extracted as the itch behavior data 32a. If there are multiple types of itch behavior, the itch behavior analysis unit 214 makes a judgment for each type of itch behavior.

Next, the itch behavior analysis unit 214 stores the itch behavior data 32a in the data storage unit 13 (S25). The itch behavior data 32a stored at this time includes the type of itch behavior, the start and end times of the itch behavior, the intensity level of the itch behavior, etc.

Next, the itch behavior analysis unit 214 determines whether the hour has been reached (S26). For example, 1:00, 2:00, etc. If the hour has been reached (S26: Yes), the itch behavior analysis unit 214 determines the intensity level for the most recent hourly unit and stores it as itch behavior data 32a in the management data storage unit 232 (S27). If the hour has not been reached (S26: No), the process proceeds to the next step without performing S27.

Next, the itch behavior analysis unit 214 determines whether 24:00 has been reached (S28). If 24:00 has been reached (S28: Yes), the itch behavior analysis unit 214 determines the intensity level for the most recent day and stores it as itch behavior data 32a in the management data storage unit 232 (S29). If 24:00 has not been reached (S28: No), the process proceeds to the next step without performing the process of S29.

Then, the process returns to S21 and is repeated. In other words, until one hour has passed, only itching behavior data 32a is generated for each predetermined time period. Then, with each passing hour, hourly itching behavior data 32a is added. Furthermore, with each passing day, daily itching behavior data 32a is added.

When itch behavior data and daily activity data are output as the same type of data, the question arises as to how to distinguish between the two. As described above, daily activities such as walking and running are continuous, whereas itch behavior is instantaneous and less continuous than walking or running. Therefore, the itch behavior analysis unit 214 focuses on the duration of candidate data including itch behavior data 32a and daily activity data, and distinguishes between itch behavior data and daily activity data based on the duration of the candidate data. This allows for appropriate extraction of itch behavior data 32a.

Furthermore, embodiment 4 achieves the same effects as embodiment 3. That is, by using machine learning, it is possible to appropriately extract candidate data, which is the desired data, from a wide variety of data.

(Embodiment 5)
A pet management system 300 according to the fifth embodiment will be described with reference to Fig. 11 and Fig. 12. The functional configuration of the pet management system 300 according to the fifth embodiment differs from that according to the third embodiment in that a bridge device 4c is provided.

The management device 4 is configured to be able to communicate with the wearable device 3. The management device 4 comprises an owner terminal 4a, a server 4b, and a bridge device 4c. The bridge device 4c is installed, for example, in the owner's home or a veterinary facility, and is configured to be able to mediate communication between the wearable device 3 and the server 4b.

The bridge device 4c is configured to be able to communicate with the wearable device 3. When the bridge device 4c approaches the wearable device 3, it can perform short-range wireless communication, such as Bluetooth (registered trademark) or NFC (registered trademark). Furthermore, the bridge device 4c is configured to be able to communicate with the server 4b via the Internet. Therefore, the server 4b is configured to be able to communicate with the wearable device 3 via the bridge device 4c.

The owner terminal 4a is owned by the owner and is configured to be able to communicate with the server 4b. Note that the owner terminal 4a may be configured to be unable to communicate with the wearable device 3, or, as in embodiment 3, may be configured to be able to communicate with the wearable device 3.

The elements of the wearable device 3, the elements of the owner terminal 4a, the elements of the server 4b, and the elements of the veterinarian terminal 5a are the same as those in embodiment 3. Note that the bridge device 4c can also be applied to the configurations of embodiments 1 and 2.

(Other embodiments)
The pet management systems 1, 100, 200, and 300 of the above embodiments may be configured without the veterinarian terminal 5a. While the management device 4 is configured to include the owner terminal 4a and the server 4b, the management device 4 may be configured to include only the owner terminal 4a, with the owner terminal 4a also having the functions of the server 4b. In this case, the owner terminal 4a would include the management data storage unit 32, 232, the itching evaluation unit 33, and the display processing unit 34 that constitute the server 4b.

Server 4b may also be configured to consist of multiple servers separated by function. For example, server 4b may include a first server that constitutes the itch behavior analysis unit 14, 214, and a second server that constitutes other elements. This can be configured arbitrarily, taking into account factors such as processing load.

In addition, in the above embodiment, the wearable device 3 is configured to be able to communicate with the server 4b via the owner terminal 4a or the bridge device 4c. Alternatively, the server 4b may be configured to be able to communicate directly with the wearable device 3 and directly acquire data stored in the wearable device 3.

Claims (20)

A wearable device (3) attached to a pet (2),
A management device (4) configured to be able to communicate with the wearable device;
Equipped with
The wearable device includes sensors (11, 12) that detect at least one of acceleration and angular velocity;
The wearable device or the management device includes an itch behavior analysis unit (214) that analyzes data (13a, 232f) detected by the sensor to generate itch behavior data (13b, 32a) representing behavior of the pet associated with itch,
The management device includes:
a management data storage unit (32, 232) for storing the itch behavior data (13b, 32a);
a display unit (23) that displays the itch behavior data stored in the management data storage unit,
The pet management system (200, 300) is configured such that the itching behavior analysis unit (214) generates the itching behavior data using a machine learning model based on features selected from statistics for each frequency band divided into multiple frequency bands for the data detected by the sensor. The pet management system (200, 300) according to claim 1, wherein the feature quantity includes at least one statistical quantity for each of a plurality of frequency bands divided into the plurality of frequency bands. The pet management system (200, 300) according to claim 1, wherein the intervals between the plurality of frequency bands are 0.5 Hz to 3 Hz. The pet management system (200, 300) of claim 1, wherein the itching behavior analysis unit (214) analyzes the data detected by the sensor to distinguish between the itching behavior data and living behavior data including behavior associated with the pet walking or running, and generates the itching behavior data that does not include the living behavior data. the sensor outputs the itch behavior data and the daily living behavior data as the same type of data;
The itch behavior analysis unit (214)
extracting candidate data including the itch-related behavior data and the daily behavior data by analyzing the data detected by the sensor;
The pet management system (200, 300) according to claim 4, wherein the itching behavior data and the living behavior data are distinguished based on the duration of the candidate data, and the itching behavior data that does not include the living behavior data is generated. The itch behavior analysis unit (214)
The pet management system (200, 300) described in claim 5 extracts the candidate data using the machine learning model based on the feature selected from statistics of each frequency band divided into multiple frequency bands for the data detected by the sensor. The pet management system (200, 300) according to any one of claims 1 to 6, wherein the itch-related behavior data includes a time period during which the pet performed the itch-related behavior, or a number of times within a set period. The pet management system (200, 300) according to any one of claims 1 to 6, wherein the itch behavior data includes an intensity level of the itch-related behavior performed by the pet. the itch behavior analysis unit generates a plurality of types of itch behavior data;
The management device further includes an itch evaluation unit (33) that performs itch evaluation processing based on the plurality of types of itch behavior data to generate an itch index value (32b) that indicates the degree of itch,
The pet management system (200, 300) according to any one of claims 1 to 6, wherein the display unit displays the itching behavior data stored in the management data storage unit and the itching index value generated by the itching evaluation unit. The pet management system (200, 300) according to claim 9, wherein the itch evaluation unit generates the itch index value by performing the itch evaluation process based on the transition of each of the plurality of types of itch behavior data. the itch evaluation unit performs the itch evaluation process based on the transition of each of the plurality of types of itch behavior data, thereby generating an inflammation progression level (32c) of dermatitis;
The pet management system (200, 300) according to claim 10, wherein the display unit displays the itching behavior data stored in the management data storage unit, the itching index value generated by the itching evaluation unit, and the inflammation progression level. The management device (4)
a server (4b) configured to be able to communicate with the wearable device (3), including the management data storage unit (32, 232), and acquiring data stored in the wearable device;
an owner terminal (4a) configured to be able to communicate with the server (4b) and having the display unit;
The pet management system (200, 300) of any one of claims 1 to 6, comprising: The management device (4)
Further, a bridge device (4c) configured to mediate communication between the wearable device and the server,
The pet management system (300) of claim 12, wherein the server obtains data stored on the wearable device via the bridge device. The owner terminal is configured to be able to communicate with the wearable device (3),
The pet management system (200) according to claim 12, wherein the server is configured to be able to communicate with the owner terminal and acquires data stored in the wearable device via the owner terminal. The wearable device (3) includes the itch behavior analysis unit (214),
13. The pet management system (200) according to claim 12, wherein the server (4b) includes the management data storage unit (32) and acquires the itching behavior data via the owner terminal. The pet management system (200, 300) described in claim 12, wherein the server (4b) is equipped with the management data storage unit (232) and the itching behavior analysis unit (214), and acquires data detected by the sensor via the owner terminal. The owner terminal further includes a behavior information input unit (24) configured to input behavior information data (32d) of the pet in association with the itching behavior data displayed on the display unit,
13. The pet management system (200, 300) according to claim 12, wherein the management data storage unit of the server stores the behavior information data input by the behavior information input unit in association with the itching behavior data. The pet management system (200, 300) of claim 12 further comprises a veterinarian terminal (5a) configured to be able to communicate with the server and to be able to view the itching behavior data stored in the management data storage unit. The owner terminal includes a medical interview input unit (25) configured to input medical interview data (32e) for a veterinarian consultation,
the management data storage unit of the server stores the medical interview data input by the medical interview input unit;
The pet management system (200, 300) according to claim 18, wherein the veterinarian terminal is configured to be able to view the medical interview data. The pet management system (200, 300) according to any one of claims 1 to 6, wherein the behavior associated with itching includes at least one of head-shaking behavior, scratching behavior, and licking behavior.