KR20230071698A - A wearable apparatus to manage health care for companion animal - Google Patents

Abstract

The present invention relates to a wearable apparatus for managing health care for companion animals According to an embodiment of the present invention, a wearable apparatus for managing health care for companion animals includes: a sensor that acquires activity data of the companion animal in predetermined time units; a memory storing the acquired activity data; and a processor that detects signs of abnormal health of the companion animal or performs weight management of the companion animal from the stored activity data using a deep learning inference model.

Description

Wearable device for health care of companion animals {A wearable apparatus to manage health care for companion animal}

The present invention relates to a wearable device for health management of a companion animal.

Companion animals are a general term for animals that live together with humans, and representative examples include dogs and cats. Recently, people's awareness of companion animals is improving, and companion animal-related industries are also growing rapidly. Accordingly, the state of health of companion animals is becoming a great concern for people, and technologies for observing the state of companion animals or managing companion animals by combining IoT technology are increasing.

In particular, in the case of companion animals, unlike humans, they cannot speak, so special care is required in health management of companion animals, and technology to manage companion animal health through a technological approach is in the spotlight.

(Patent Document 01) Korean Patent Publication No. 10-2018-0067107 (published on June 20, 2018)

The present disclosure provides a wearable device for health management of a companion animal.

The problem to be solved by the present invention is not limited to the above-mentioned problems, and other problems and advantages of the present invention that are not mentioned can be understood by the following description and more clearly understood by the embodiments of the present invention. It will be. In addition, it will be appreciated that the problems and advantages to be solved by the present invention can be realized by the means and combinations indicated in the claims.

A first aspect of the present disclosure is a wearable device for health management of a companion animal, comprising: a sensor that acquires activity data of a companion animal in units of a predetermined time; a memory for storing the acquired activity data; and a processor configured to detect signs of abnormal health of the companion animal or manage the weight of the companion animal from the stored activity data using a deep learning inference model.

Here, the deep learning inference model may include at least one of a C-LSTM inference model and a GRU inference model.

The processor infers the daily life pattern of the companion animal from the activity data using the deep learning inference model, and detects signs of abnormality of the companion animal or manages the weight of the companion animal based on the daily life pattern. can be done

The processor determines the number of repetitions of a specific behavior of the companion animal estimated from the activity data, the content of the first activity of the companion animal by analyzing the aspect of the specific behavior, the first daily life pattern of the companion animal, and the same breed. By comparing at least two or more of the second activity contents corresponding to the normal activity range of the companion animal, it is possible to detect signs of abnormal health of the companion animal.

The processor determines the first activity amount of the companion animal estimated from the activity data, the daily life pattern of the companion animal, and the second activity amount corresponding to the normal activity range of the normal weight group of the same breed or the normal activity of the overweight group. The risk of obesity of the companion animal may be predicted by comparing at least two of the third activity amounts corresponding to the range.

The processor may provide an activity log of the companion animal including at least one of a result of analyzing the activity of the companion animal and a result of analyzing an activity amount of the companion animal to the user through an application.

The processor may provide a warning notification to the user when a symptom of a health abnormality is detected for the companion animal or an obesity prediction result of the companion animal exceeds a predetermined range.

In addition, a communication module supporting at least one communication method among Wi-Fi, Bluetooth, and mobile communication may be further included.

The processor may include adaptively selecting at least one communication method among communication methods supported through the communication module.

The processor may further include a GPS sensor that receives a GPS signal, and the processor estimates a current location of the device based on the GPS signal, and the communication method according to the current location of the device and a battery management scenario according to a situation. may include selecting

A second aspect of the present disclosure may be a computer-readable recording medium recording a program for executing a wearable device for health management of a companion animal on a computer.

Companion animal health can be managed through a wearable device that is suitable for companion animals and records the daily life of companion animals.

By combining IoT technology and deep learning technology, it is possible to detect signs of health abnormalities that may occur in companion animals more quickly. This may be a sign of health abnormalities that a person may not be aware of in advance, and these problems can be solved through a technological approach.

Companion animal activity data can be analyzed using a deep learning inference model, and companion animal activity logs can be shared through companion animal wearable devices to enhance understanding of companion animals.

By selecting the LSTM inference model or the GRU inference model among deep learning models, it is possible to learn about the behavior patterns of companion animals, and by using the information obtained in the previous step as an input for the inference model, it is possible to detect abnormal symptoms tailored to companion animals. can

1 and 2 are examples of a wearable device for health management of companion animals and a method for managing health of companion animals using the wearable device according to an embodiment of the present invention.
3 is an example of a process of acquiring activity data of a companion animal and estimating a specific behavior of the companion animal from the obtained activity data according to an embodiment of the present invention.
4A to 4D are examples of estimating a specific behavior of a companion animal based on animal activity data according to an embodiment of the present invention. 5 is an example showing a frequency of a specific behavior estimated for a companion animal according to an embodiment of the present invention.
6 is a diagram explaining detection of symptoms of abnormal health of a companion animal using a deep learning inference model according to an embodiment of the present invention. 7 is a diagram explaining predicting obesity risk of a companion animal using a deep learning inference model according to an embodiment. 8 is a table describing battery management scenarios of a wearable device according to an embodiment of the present invention.
9 is an example of a wearable device according to an embodiment of the present invention.

Advantages and features of the present invention, and methods for achieving them will become clear with reference to the detailed description of embodiments in conjunction with the accompanying drawings. However, it should be understood that the present invention is not limited to the embodiments presented below, but may be implemented in a variety of different forms, and includes all conversions, equivalents, and substitutes included in the spirit and scope of the present invention. . The embodiments presented below are provided to complete the disclosure of the present invention and to fully inform those skilled in the art of the scope of the invention to which the present invention belongs. In describing the present invention, if it is determined that a detailed description of related known technologies may obscure the gist of the present invention, the detailed description will be omitted.

Terms used in this application are only used to describe specific embodiments, and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this application, the terms "include" or "have" are intended to designate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, but one or more other features It should be understood that the presence or addition of numbers, steps, operations, components, parts, or combinations thereof is not precluded.

Some embodiments of the present disclosure may be represented as functional block structures and various processing steps. Some or all of these functional blocks may be implemented as a varying number of hardware and/or software components that perform specific functions. For example, functional blocks of the present disclosure may be implemented by one or more microprocessors or circuit configurations for a predetermined function. Also, for example, the functional blocks of this disclosure may be implemented in various programming or scripting languages. Functional blocks may be implemented as an algorithm running on one or more processors. In addition, the present disclosure may employ prior art for electronic environment setup, signal processing, and/or data processing, etc. Terms such as “mechanism,” “element,” “means,” and “composition” will be used broadly. It can be, and is not limited to mechanical and physical configurations.

In addition, connecting lines or connecting members between components shown in the drawings are only examples of functional connections and/or physical or circuit connections. In an actual device, connections between components may be represented by various functional connections, physical connections, or circuit connections that can be replaced or added.

Hereinafter, the present disclosure will be described in detail with reference to the accompanying drawings.

1 and 2 are examples of a wearable device 10 for health management of companion animals and a method for managing health of companion animals using the wearable device 10 according to an embodiment of the present invention.

Here, the device 10 may have a form of a wearable device that can be attached to a leash, collar, clothes, bag, or the like of a companion animal. In this case, the device 10 may be manufactured to a size that does not interfere with the companion animal's activities and be attached to the companion animal's body or clothing.

Referring to FIG. 1 , the device 10 may include a sensor 12 , a memory 13 , a processor 14 , and a communication module 15 .

The sensor 12 may acquire activity data of a companion animal in units of a predetermined time. Here, the sensor 12 may include an acceleration sensor, a gyro sensor, or a GPS sensor.

For example, an acceleration sensor is a sensor that detects a change in speed and can detect size information measured from three vectors having x, y, and z axes. For example, when the sensor 12 includes an acceleration sensor, the sensor 12 may collect raw-data corresponding to acceleration information 10 to 100 times per second. More preferably, sensor 12 can collect raw data 30 to 70 times per second. More preferably, sensor 12 can collect raw data 50 times per second. Compared to the prior art, the sensor 12 acquires raw data more frequently per predetermined unit of time (eg, 1 second). Accordingly, the device 10 or the server may more accurately infer the daily life pattern of the companion animal compared to the prior art.

As another example, the gyro sensor is a sensor that detects an angular velocity and may collect rotation angle information of a companion animal. For example, when the sensor 12 includes a gyro sensor, the sensor 12 may measure data about the movement direction of the companion animal in units of seconds from the angular velocity.

Meanwhile, the sensor 12 may collect activity data of the companion animal at predetermined time units. For example, the sensor 12 may include an acceleration sensor or a gyro sensor, a combination thereof, or other types of sensors.

The GPS sensor may receive location information indicating the current location of device 10 from satellites. When a companion animal equipped with the device 10 performs an external activity such as a walk, the location of the companion animal may be tracked through a GPS signal, and the device 10 may provide location information of the companion animal to the user. Meanwhile, indoors where the GPS reception signal is weak, the location of the companion animal may be collected through a different positioning method using a communication module.

The memory 13 may store acquired activity data. Activity data of the companion animal collected from the sensor may be cumulatively recorded through the memory 13 . In addition, the memory 13 may record activity data through the processor 14 and a specific behavior or daily life pattern of the companion animal estimated through a deep learning inference model. In addition, the memory 13 may record a companion animal activity log in which information collected on a daily life pattern or specific behavior of the companion animal is processed through the processor 14 .

The processor 14 may detect signs of abnormal health of the companion animal or manage the weight of the companion animal from the stored activity data using a deep learning inference model.

Here, there may be several types of deep learning inference models. For example, the deep learning inference model may include a recurrent neural network (RNN), a convolutional long short-term memory (C-LSTM), and a gated recurrent unit (GRU), which are models that deal with sequential data over time.

In order for the reasoning model of the present invention to detect the companion animal's daily life pattern and health abnormal signs from the companion animal's activity data or to manage the companion animal's weight, the RNN structure that inputs the previous time step as the state in the next layer It may be desirable to select a deep learning model of However, although the deep learning model of the RNN structure has several advantages, some problems may occur when long-term log records are input to the inference model. In order to solve this problem, the present invention can detect abnormal signs of a companion animal mainly using C-LSTM or GRU or an inference model combining them.

For example, selecting an optimal inference model among deep learning inference models can stabilize the performance of the device 10 and obtain better inference results. In addition, C-LSTM or GRU models show better efficiency in terms of gradient loss, congestion, and long term dependencies than RNN models.

According to an embodiment of the present invention, the device 10 may select a C-LTSM model and/or a GRU model as a deep learning inference model. The C-LSTM inference model can perform inference dependent on a predetermined time by adding the cell state of the previous layer to the hidden state vector at the time step of each layer. In addition, the device 10 may erase some of the memories up to the previous cell by adding a forget gate in some layers in the C-LSTM inference model. Due to this, the device 10 can prevent function congestion due to an increase in the amount of data to be processed when using the C-LSTM inference model and play a role of containing meaningful information.

Meanwhile, the GRU model has a more standardized structure than the C-LTSM model and can be calculated quickly. Unlike the LSTM inference model that used cell state and hidden state vectors, only hidden state vectors can be used.

According to an embodiment, the processor 14 may infer a daily life pattern of a companion animal from activity data using a deep learning inference model (210). Meanwhile, the deep learning inference model used in the present invention may be referred to as a "NeuroSense Algorithm".

For example, the processor 14 may learn an activity data accumulation record for a predetermined long time collected from the sensor 12 as learning data. At this time, the activity data may have the form of a graph in which the magnitude of change in speed over time is recorded, and the processor 14 stores the activity data in units of time or daily to generate raw data for inferring daily life patterns. data can be processed.

The processor 14 may output an intermediate estimation result obtained by analyzing the number of repetitions of a specific behavior of the companion animal and aspects of the specific behavior through the neurosense algorithm as the content of the first activity of the companion animal. For example, the processor 14 may infer a daily life pattern of a companion animal that is repeated for a predetermined time based on the first activity through a neurosense algorithm and output it as an intermediate estimation result.

In addition, the processor 14 may detect signs of abnormality of the companion animal based on the daily life pattern (220).

For example, the processor 14 may determine a predetermined action corresponding to the activity data of the companion animal using a neurosense algorithm. In addition, the processor 14 may identify a specific behavior related to the health abnormality among the determined predetermined behaviors. In addition, the processor 14 determines the number of repetitions of a specific behavior of the companion animal estimated from the activity data, the first activity content of the companion animal by analyzing the aspect of the specific behavior, the first daily life pattern of the companion animal, and the same breed. By comparing at least two or more of the second activity contents corresponding to the normal activity range of the companion animal, abnormal symptoms related to the health of the companion animal may be detected.

For example, the processor 14 refers to the companion animal's daily life pattern, and if the companion animal performs a specific scratching behavior more than a threshold value than usual, the processor 14 identifies an abnormal symptom related to the companion animal's health (for example, skin disease). suspicious signs).

As another example, the processor 14 may refer to the daily life pattern of the companion animal, and detect a lot of tossing in sleep behavior or frequent waking up without deep sleep as a symptom of abnormal health of the companion animal. In this case, even when the device 10 cannot accurately infer the reason for the change in the behavior pattern of the companion animal, it can notify the user that an “abnormal symptom” has been found. If such “abnormal signs” are continuously detected for a certain period of time, the device 10 may send countermeasures or recommendations to the user together.

On the other hand, the companion animal's scratching behavior or licking behavior may not be unusual compared to the companion animal's daily life pattern or the daily life pattern of other companion animals. The NeuroSense algorithm may further consider the second activity content corresponding to the normal activity range of the same breed in order to produce an inference result more suitable for the corresponding companion animal.

For example, the NeuroSense algorithm may use a database related to activity contents corresponding to a normal activity range for each breed and age of a companion animal. It can be noted that more activity data may be collected if the companion animal is young, and less activity data may be collected if the companion animal is older. For example, the neurosense algorithm may apply the companion animal's age as an addition or subtraction factor to set a threshold for classifying the companion animal's specific behavior.

According to another embodiment, the processor 14 may manage the weight of the companion animal based on the daily life pattern (220). The processor 14 may receive activity data, type, age, and weight-related information as companion animal information in order to manage the weight of the companion animal. In this case, the device 10 may predict the obesity risk of the companion animal even if some of the companion animal information is not present. For example, even if weight information of the companion animal is not input by the user, the apparatus 10 may predict the obesity risk of the companion animal based on the age and breed of the companion animal.

Here, the processor 14 may use the GRU as an inference model for the neurosense algorithm. For example, the processor 14 learns in advance a second normal activity range corresponding to a normal weight group of the same breed as the companion animal or a third normal activity range corresponding to an overweight group of the same breed, and uses this to It is possible to determine the appropriate weight of a specific companion animal or to predict the risk of obesity of a companion animal.

For example, the processor 14 calculates the difference between the accumulated activity data of the companion animal and the normal activity amount of the normal/overweight group within the same breed, and reflects this as a factor in the GRU Cell to predict the obesity risk of the companion animal. can A detailed description of this will be described later with reference to FIG. 5 .

Meanwhile, the processor 14 uploads intermediate estimation results or output data using the neurosense algorithm to the server, so that the corresponding contents can be shared with the user. Here, the server may be a server that runs an application that works with the wearable device 10, and the application provides the user with the result of analyzing the activity of the companion animal, the result of analyzing the amount of activity, or the daily life pattern information of the companion animal. can In this case, the application may process the data output through the deep learning inference model in the form of a companion animal activity log and provide the data to the user.

For example, a user may obtain information about companion animal activity data, an activity log, a daily life pattern, an appropriate weight, an obesity risk prediction value, and health abnormal signs from the device 10 through an application. In addition, if a sign of abnormal health of the companion animal is detected, the device 10 may further provide a warning notification to the user through an application.

The communication module 15 supports at least one communication method among Wi-Fi, Bluetooth, and mobile communication, and the processor 14 can adaptively select a communication method supported through the communication module according to circumstances. Here, mobile communication may include an eMTC communication scheme supporting LTE.

For example, the processor 14 may estimate the current location of the device 10 based on the received GPS signal, and determine the communication mode change according to the current location of the device 10 and battery management scenarios according to circumstances. . A detailed description of the battery management scenario will be described later with reference to FIG. 8 .

3 is an example of a process of acquiring activity data of a companion animal and estimating a specific behavior of the companion animal from the obtained activity data according to an embodiment of the present invention.

First, device 10 may obtain activity data ( 310 ). Here, the activity data may be compared on the same plane by plotting the activity data with magnitude values obtained from several types of sensors. For example, the companion animal may acquire x-axis, y-axis, and z-axis activity data from a sensor attached to the device 10 .

Device 10 may then store the acquired activity data in memory (320).

Next, the device 10 may perform behavior estimation of the companion animal using the neurosense algorithm (330). Here, the behavior estimation may be estimating the companion animal's behavior corresponding to the acquired activity data, and may further include estimating a specific behavior that may be an indicator of a health abnormality among the companion animal's behaviors (340). ). For example, the device 10 may classify specific behaviors as several indicators for estimating the companion animal's health abnormality among numerous behaviors of the companion animal.

For example, when the device 10 selects sleep behavior, licking behavior, scratching behavior, water drinking behavior, etc. as a specific activity as an index informing of a companion animal's health abnormality, the device 10 determines whether the specific behavior is estimated. And the frequency and pattern of specific behaviors can be analyzed.

For example, the processor 14 may classify the sleep behavior of the companion animal into Restful, Slightly Disrupted, and Disrupted using the classifier model.

As another example, the processor 14 may count the number of licking behaviors or scratching behaviors of the companion animal. For example, the processor 14 uses a classifier model to determine frequent, occasional, elevated, and severe behaviors according to the number of licking or scratching behaviors performed by the companion animal during the day. etc. can be classified.

As another example, the processor 14 compares the water drinking behavior of the companion animal with a predetermined average value using the classifier model and classifies it into Below Average, Average, Above Average, and the like. can

4A to 4D are examples illustrating estimating a behavior of a companion animal based on activity data of the companion animal according to an embodiment of the present invention.

According to an example, the activity data obtained through the device 10 may be specifically learned according to a pattern of a companion animal's behavior. In this case, an image of a companion animal may be used to match a specific behavior of the companion animal with a pattern of activity data. For example, recognizing a specific behavior of a companion animal from an image of a companion animal, and obtaining a specific behavior of a companion animal and matching activity data of the companion animal through time synchronization with activity data acquired from the device 10 can do. Referring to FIG. 4A , the device 10 displays activity data of the x-axis, y-axis, and z-axis collected through a plurality of sensors in an overlapping manner, and through the NeuroSense algorithm, the pattern represented by each graph and the companion animal's behavior can be matched. As an example of FIG. 4A , it can be estimated that the companion animal performed actions such as shaking (401), walking (402), running (403), and shaking (404) for a predetermined period of time.

Referring to FIG. 4B , since activity data acquired for each puppy may be collected differently according to the behavioral aspect of a specific puppy, the activity data obtained from puppy A and the activity data obtained from puppy B are different even for the same shaking behavior. patterns can be seen.

On the other hand, referring to FIG. 4B , dog A's shaking behavior (Shaking, 411) and dog B's shaking behavior (Shaking, 412) may not be accurately distinguished by the human eye, but the NeuroSense algorithm It can be distinguished by learning the behavior pattern for . At this time, the NeuroSense algorithm can be used as an addition or subtraction factor in estimating the behavior of each companion animal according to the type and age of the companion animal.

Similarly, referring to FIGS. 4C to 4D , FIG. 4C illustratively shows activity data related to puppy A's scratching behavior (Scratching, 431) and puppy B's scratching behavior (Scratching, 432). As another example, FIG. 4D illustratively shows activity data related to drinking and eating behaviors (Drinking and Eating 441) of puppy A and drinking and eating behaviors (Drinking and Eating 442) of puppy B.

5 is an exemplary diagram illustrating a frequency of a specific behavior of a companion animal estimated according to an embodiment of the present invention. Referring to FIG. 5 , the companion animal's daily life pattern may be inferred from the behavior of the companion animal estimated for a predetermined period of time. For example, referring to FIG. 5 , the corresponding companion animal has a high frequency of running behavior. In addition, the frequency of the companion animal's walking behavior is also at a middle-to-high level. Here, the degree of high/low or high/middle/low levels of the behavioral frequency may be presented as a relative value compared to the average behavioral frequency of other companion animals.

If a companion animal has very little activity such as walking or running, or if the activity of a companion animal, which was previously abundant, rapidly decreases, the device 10 detects the specific behavior of the companion animal. It is possible to detect symptoms of abnormalities in companion animals from the activity frequency for .

FIG. 6 is a diagram explaining detection of symptoms of abnormal health of a companion animal using a neurosense algorithm according to an embodiment of the present invention.

Referring to FIG. 6, the data handled through the C-LSTM inference model of the NeuroSense algorithm can be largely described as follows in a learning step, an input step, an intermediate output step, and an output step.

In the input step, the device 10 may input activity data of the companion animal as input data 410 to the C-LSTM inference model. Here, the device 10 performs a predetermined preprocessing of intermediate processing with several inputs according to a predetermined time step in order to input the input (INPUT) value to each layer of the LSTM-inference model using the C-LSTM inference model. can be done

In the learning step, the learning data 420 may include a cumulative record of activity data and content of a second activity corresponding to a normal activity range of the same breed. Here, the activity data accumulation record may be a record in which activity data collected in units of a predetermined time is accumulated. The accumulated record of companion animal activity data can be input to the hidden layer of the C-LSTM inference model or used to infer the companion animal's repeated daily life patterns.

For example, when activity data for a companion animal is collected over several days or months, the processor 14 discovers a pattern related to estimated behavior of the companion animal through a C-LSTM inference model, and finds a pattern for the corresponding companion animal. can be set to the normal activity range of In addition, the processor 14 may monitor changes in the daily life pattern of the companion animal through activity data newly added after predetermined learning.

In addition, the processor 14 may learn an inference model having a characteristic value for each behavior corresponding to a normal activity range of the same breed as a parameter. For example, the processor 14 builds a breed-specific database for each companion animal belonging to the same breed, collects the companion animal's activity data and the animal's behavior corresponding to the corresponding activity data, and uses a C-LSTM model in a supervised learning method. You can build a learning data-set for learning.

Here, when the learning data-set is built, the processor 14 groups the main variable and the dependent variable to calculate the correlation between each variable, which is used by the C-LSTM inference model to detect signs of abnormal health of the companion animal. It can be set as a parameter for

The C-LSTM inference model is a kind of artificial neural network model, and can be used to output intermediate estimation results 430 and output data 440 through the following steps.

For example, the intermediate estimation result 430 may be the first activity content of the companion animal, such as the first daily life pattern of the companion animal, the estimation of a specific action of the companion animal, the number of repetitions of a specific action, and aspect analysis.

For example, the C-LSTM inference model can extract N-gram features through convolution. As an example, a one-dimensional convolution product may include a filter vector that slides over the sequence and detects features at different locations. Through this, n feature maps generated for n filters having the same length may be rearranged into feature expressions for each window.

In addition, the C-LSTM inference model performs maximum overpooling or dynamic K maximum pooling through Max-Pooling, and among the features rearranged through convolution, features related to specific behaviors required to detect signs of abnormal health of companion animals. can choose

A C-LSTM inference model can have at least two or more layers, and the cell state in each layer can be input as an input to the next layer. For example, an intermediate output may be input as a vector in a first layer and a subsequent layer of C-LSTM.

Each layer of C-LSTM receives the previous operation result as an input to the next layer, and some layers may include a forget gate to prevent overfitting of the inference model. The FC Layer (Fully Connected Layer) may perform linear conversion according to preset parameters and output an abnormal sign detection result for the companion animal as a probability as a final result.

In this case, various types of classifiers (SVMs) may be used in the C-LSTM model, and a step of setting a variable to be applied with priority in each layer and an appropriate weight for each variable may be included.

In the output stage, the C-LSTM inference model can detect signs of abnormal health of companion animals with output data 440 through at least two or more LSTM layers and a final layer (FC Layer).

Here, when the activity data for the companion animal is input, the processor 14 estimates the behavior of the companion animal from the activity data through the C-LSTM inference model, and among the behaviors of the companion animal, specific behaviors related to symptoms of abnormal health It is possible to analyze the behavioral patterns of companion animals' behaviors by scoring them and calculating scores for each behavior. In addition, the analyzed activity details (second activity details) of the companion animal may be compared with a normal activity range by comparing with a predetermined threshold value.

For example, if several behaviors are estimated from activity data in the first step through the C-LSTM model, in the second step, among the estimated behaviors in the first step, specific behaviors and behavioral patterns related to health abnormal symptoms are classified. and the content of the first activity can be analyzed. Then, based on the contents of the first activity analyzed in the second step, symptoms of abnormal health of the companion animal may be detected.

7 is a diagram explaining prediction of obesity risk of a companion animal using a neurosense algorithm according to an embodiment. Here, as an example of the neurosense algorithm, the determination of the appropriate weight of the companion animal and the prediction of the obesity risk of the companion animal through the GRU Cell will be described.

Referring to FIG. 7 , an input step, a learning step, and an output step (including an intermediate estimation result) in which each input data is input in parallel using a GRU Cell can be described as follows.

In the input step, the processor 14 may acquire weight and activity data of a specific companion animal and input them into the neurosense algorithm as input data 510 . At this time, the input data 510 is companion animal information, and may include activity data, the type of companion animal, age, weight, and the like. The processor 14 may combine at least one piece of the above information or a plurality of pieces of information and input the information to the neurosense algorithm.

The processor 14 may analyze the amount of activity of the companion animal from the activity data of the companion animal input through the neurosense algorithm. Here, the activity data may be used as a hidden state vector by inputting a plurality of activity data collected in different time ranges in parallel to each of the GRU Cells according to a predetermined time unit.

In the learning step, the learning data 520 includes a cumulative record of activity data, a second activity amount corresponding to the normal activity range of the normal weight group of the same breed, and/or a third activity amount corresponding to the normal activity range of the overweight group of the same breed. can include

Here, activity data for several breeds of companion animals rather than specific companion animals may be collected in advance. For example, a database including activity data collected for companion animals of the same breed as companion animals and corresponding second activity amounts and/or third activity amounts as a learning data-set may be constructed.

In the GRU Cell, each state of input data can be input as a hidden state vector in the next step. Referring to FIG. 5, the GRU Cell may calculate a final output through a predetermined intermediate estimation step.

Here, the intermediate estimation result 530 is a result estimated in an intermediate process for determining the appropriateness of the companion animal's weight and predicting the obesity risk of the companion animal, and can be made at any stage of the GRU Cell. For example, the intermediate estimation result 530 may include a first activity amount of the companion animal estimated from activity data and a first daily life pattern of the companion animal.

In the final output step, the processor 14 may output the result of determining the appropriateness of the weight of the companion animal and the prediction result of the obesity risk of the companion animal as output data 540 . The obesity risk prediction result may be presented as a probability value or an indicator corresponding to the probability value.

Here, the processor 14 selects a second activity range corresponding to a normal activity range of a normal weight group among other companion animal groups of the same type as the first activity amount of the companion animal estimated from the acquired activity data using a neurosense algorithm. By comparing the activity level or the third activity level corresponding to the normal activity range of the overweight group, it is possible to determine the appropriateness of the weight of the specific companion animal and to predict the obesity risk of the specific companion animal.

8 is a table describing battery management scenarios of a wearable device according to an embodiment of the present invention. Referring to FIG. 6 , the device 10 may perform battery management by selecting an appropriate communication method according to circumstances.

For example, when a companion animal wearing the device 10 is with a guardian, the device 10 prioritizes a Bluetooth connection as a communication method with the guardian's terminal (ie, the terminal currently possessed by the guardian). You can choose. At this time, when the device 10 communicates with the guardian's terminal only in the Bluetooth method, the lifespan (hereinafter referred to as 'battery life') when the battery of the device 10 is fully charged may be about 4 months, Not limited to this.

Similarly, when a companion animal wearing the device 10 is at home, Bluetooth and/or eMTC (LTE Cellular) may be applied as a communication mode between the guardian's terminal and the device 10, in which case, in units of 2 to 5 minutes. As a result, communication may be performed between the guardian's terminal and the device 10 . In this case, the battery life of the device 10 may be about 3 months, but is not limited thereto.

In addition, when the companion animal wearing the device 10 is outside, the device 10 may use any one of Bluetooth and eMTC as a communication method with the guardian's terminal. At this time, communication may be performed between the guardian's terminal and the device 10 every 2 to 5 minutes. At this time, the battery life of the device 10 may be about one month, but is not limited thereto.

Meanwhile, when the companion animal is missing, the device 10 may be switched to a rescue mode through an application of the guardian terminal. At this time, the device 10 may receive a GPS signal in units of 1 minute for location tracking based on the GPS signal. Also, in this case, the eMTC method may be used together for communication between the device 10 and the application. At this time, the battery life of the device 10 may be about 2 days, but is not limited thereto.

On the other hand, the communication method described in the above example is according to a plurality of communication modes and battery usage scenarios that can be selected by the device 10, and in some cases, Wi-Fi, cellular, short-range wireless communication, etc. Communication methods may be used interchangeably.

9 is an example of a wearable device according to an embodiment of the present invention. The device 10 of the present invention is a kind of smart wearable device, and can track the companion animal's daily life and provide a companion animal's activity log to the user.

At this time, through an application that works with the device 10, the data collected and analyzed by the device 10 is directly provided to the user through a communication method such as Bluetooth or uploaded to a server that manages the application through Wi-Fi. By doing so, various information can be provided to the user.

Meanwhile, the present invention may be implemented as a computer-readable recording medium recording a program for executing the wearable device 10 for health management of a companion animal on a computer.

According to an embodiment of the present invention, the method according to various embodiments of the present disclosure may be provided by being included in a computer program product. Computer program products may be traded between sellers and buyers as commodities. A computer program product is distributed in the form of a device-readable storage medium (eg compact disc read only memory (CD-ROM)), or via an application store (eg Play Store™) or between two user devices. It can be distributed (e.g., downloaded or uploaded) directly or online. In the case of online distribution, at least part of the computer program product may be temporarily stored or temporarily created in a storage medium readable by a device such as a manufacturer's server, an application store server, or a relay server's memory.

The steps may be performed in any suitable order, provided that the order of steps constituting the method according to the present invention is not explicitly stated or stated to the contrary. The present invention is not necessarily limited by the order of description of the steps. The use of all examples or exemplary terms (eg, etc.) in the present invention is simply to explain the present invention in detail, and the scope of the present invention is limited due to the above examples or exemplary terms that are not limited by the claims. It is not. In addition, those skilled in the art can appreciate that various modifications, combinations and changes can be made according to design conditions and factors within the scope of the appended claims or equivalents thereof.

The spirit of the present invention should not be limited to the described embodiments and should not be defined, and all ranges equivalent to or equivalent to this claim as well as the following claims belong to the scope of the spirit of the present invention. something to do.

10: Wearable device for health management of companion animals
12: sensor
13: memory
14: Processor
15: communication module

Claims (1)

In the wearable device for health management of companion animals,
A sensor that acquires activity data of a companion animal in units of predetermined time;
a memory for storing the acquired activity data; and
A processor that detects signs of abnormal health of the companion animal or manages the weight of the companion animal from the stored activity data using a deep learning inference model.

Images