JP7049030B1 - Animal health management system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable a breeder to grasp the health condition of an animal at an early stage.
SOLUTION: This is a system for managing the health of an animal, and the answer to the first question related to the health condition of the animal is used as input data, and the first question to be asked after the first question designated by a veterinary medical worker. A learning model storage unit that stores a learning model created by machine learning using the second question as teacher data, a question transmission unit that sends the first question to the owner terminal, and an answer to the first question is received from the owner terminal. It is characterized by including an answer receiving unit and an inference unit that gives an answer to a learning model and asks for a second question.
[Selection diagram] Fig. 1

Description

The present invention relates to an animal health management system.

Interviews are conducted via a network (see Patent Document 1).

International Publication No. 2004/104880

Interviews are generally given when the patient feels a subjective symptom. However, in veterinary medicine, the breeder needs to know the health condition.

The present invention has been made in view of such a background, and an object of the present invention is to provide a technique capable of grasping the health condition of an animal at an early stage by a breeder.

The main invention of the present invention for solving the above-mentioned problems is a system for managing the health of an animal, and the answer to the first question related to the health condition of the animal is used as input data and designated by a veterinary medical worker. A learning model storage unit that stores a learning model created by machine learning using the second question to be asked after the first question as teacher data, and an inquiry transmission unit that transmits the first question to the owner terminal. It is characterized by including an answer receiving unit that receives the answer to the first question from the owner terminal, and an inference unit that gives the answer to the learning model and asks for the second question.

Other problems disclosed in the present application and solutions thereof will be clarified by the columns and drawings of the embodiments of the invention.

According to the present invention, the breeder can grasp the health condition of the animal at an early stage.

It is a figure which shows the whole configuration example of the periodical examination system of this embodiment. It is a figure which shows the hardware configuration example of management server 20. It is a figure which shows the software configuration example of management server 20. It is a figure explaining the operation of the periodical examination system of this embodiment.

<Outline of the invention>
The contents of the embodiments of the present invention will be described in a list. The present invention includes, for example, the following configuration.
[Item 1]
A system that manages the health of animals
The answer to the first question related to the health condition of the animal was used as input data, and the second question to be asked after the first question specified by the veterinary medical worker was created by machine learning as teacher data. A learning model storage unit that stores the learning model,
The medical examination transmission unit that sends the first question to the owner's terminal,
An answer receiving unit that receives the answer to the first question from the owner terminal, and an answer receiving unit.
An inference unit that gives the answer to the learning model and asks for the second question,
An animal health management system characterized by being equipped with.
[Item 2]
The animal health management system according to item 1.
The learning model storage unit stores the learning model for each animal species, and stores the learning model.
The reasoning unit gives the answer to the learning model corresponding to the animal species of the animal.
An animal health management system featuring.
[Item 3]
The animal health management system described in item 2,
The learning model storage unit stores the learning model in association with at least one of the animal species, breed, weight, body weight rank, age and age.
The reasoning unit provides the answer to the learning model corresponding to the animal species of the animal and at least one of the breed, the weight, the weight rank, the age and the age of the animal.
An animal health management system featuring.
[Item 4]
The animal health management system according to item 1.
The learning model is learned by using the answer and the animal species as the input data.
The reasoning unit provides the learning model with the animal species of the animal and the answer.
An animal health management system featuring.
[Item 5]
The animal health management system according to item 4.
The learning model is obtained by learning the answer, the animal species, and at least one of a breed, a body weight, a body weight rank, an age, and an age as the input data.
The reasoning unit gives the learning model the animal species of the animal, the answer, and at least one of the breed, the weight, the weight rank, the age, and the age of the animal.
An animal health management system featuring.
[Item 6]
The animal health management system according to any one of items 1 to 5.
The learning model is learned by further adding the health condition of the animal determined by the veterinary medical worker to the input data.
Further provided with a health condition acquisition unit that outputs the answer to the veterinary medical worker and accepts the input of the health condition from the veterinary medical worker.
The reasoning unit further gives the input health state to the learning model.
An animal health management system featuring.
[Item 7]
The animal health management system according to any one of items 1 to 6.
The learning model was learned by further adding the test results for the animal to the input data.
Further provided with an inspection result acquisition unit for acquiring the inspection result for the animal,
The inference unit further gives the acquired test result to the learning model.
An animal health management system featuring.
[Item 8]
The animal health management system according to any one of items 1 to 7.
Further equipped with an answer history storage unit for storing the answer received from the owner terminal,
The learning model is learned by including the answer including the time lag in the input data.
The reasoning unit gives the learning model the answer stored in the answer history storage unit and the time lag from the present time.
An animal health management system featuring.

<Overview of the system>
Hereinafter, a periodical examination system according to an embodiment of the present invention will be described. The periodical examination system of the present embodiment is intended to support the periodical examination of animals.

The periodic examination system of the present embodiment attempts to determine the health condition of an animal, especially the necessity of a visit (triage), by conducting regular interviews with the owner of an animal such as a pet from the veterinary medical institution side. It is a thing. In this embodiment, the content of the interview can be automatically determined using AI. The AI can be, for example, a classifier that determines the question to be asked next among the questions prepared in advance. For AI, a learning model in which a question manually determined by a veterinary medical worker such as a veterinarian is learned can be used. Further, in the periodical examination system of the present embodiment, the learning model is created for each animal species, but it is also possible to give the animal species as input data when learning AI. In addition, in learning AI, the owner can give an image of a specific part of the animal as input data.

FIG. 1 is a diagram showing an overall configuration example of the periodical examination system of the present embodiment. The periodic examination system of the present embodiment includes a management server 20. The management server 20 is communicably connected to each of the owner terminal 10 and the medical institution terminal 30 via the communication network 40. The communication network 40 is, for example, the Internet, and is constructed by a public telephone line network, a mobile phone line network, a wireless communication path, Ethernet (registered trademark), or the like.

The owner terminal 10 is a computer operated by an animal caregiver (owner or caregiver). The owner terminal 10 is, for example, a smartphone, a tablet computer, a personal computer, or the like. It is assumed that the owner terminal 10 is provided with a camera (not shown) and is capable of taking pictures.

The medical institution terminal 30 is a computer operated by a veterinary medical person (a nurse, a doctor, or the like in a veterinary hospital) at a veterinary medical institution. The medical institution terminal 30 is, for example, a tablet computer. The medical institution terminal 30 can be any computer such as a smartphone or a personal computer.

The management server 20 is a computer that conducts regular interviews and determines whether or not an animal patient needs to visit the hospital. The management server 20 may be a general-purpose computer such as a workstation or a personal computer, or may be a virtual computer logically realized by cloud computing.

FIG. 2 is a diagram showing a hardware configuration example of the management server 20. The configuration shown in the figure is an example, and may have other configurations. The management server 2 includes a CPU 201, a memory 202, a storage device 203, a communication interface 204, an input device 205, and an output device 206. The storage device 203 stores various data and programs, such as a hard disk drive, a solid state drive, and a flash memory. The communication interface 204 is an interface for connecting to the communication network 3, for example, an adapter for connecting to Ethernet (registered trademark), a modem for connecting to a public telephone network, and a wireless communication device for performing wireless communication. , USB (Universal Serial Bus) connector for serial communication, RS232C connector, etc. The input device 205 is, for example, a keyboard, a mouse, a touch panel, a button, a microphone, or the like for inputting data. The output device 206 is, for example, a display, a printer, a speaker, or the like that outputs data. Each functional unit of the management server 20, which will be described later, is realized by the CPU 201 reading a program stored in the storage device 203 into the memory 202 and executing the program, and each storage unit of the management server 20 is the memory 202 and the storage device 203. Is realized as part of the storage area provided by.

<Software configuration of management server 20>
FIG. 3 is a diagram showing a software configuration example of the management server 20. The management server 20 includes an inquiry transmission unit 211, an answer reception unit 212, a visit necessity transmission unit 214, a visit necessity acquisition unit 215, a learning processing unit 216, an owner information storage unit 231 and an animal information storage unit 232, and a question storage unit 233. , Questionnaire storage unit 234, answer history storage unit 235, triage history storage unit 236, and learning model storage unit 237.

The owner information storage unit 231 stores information (hereinafter referred to as owner information) regarding the owner of an animal (patient animal) who visits a veterinary medical institution. The owner information includes the owner ID, name, address, and contact information that identify the owner. It may include information about the owner other than name, address and contact information.

The animal information storage unit 232 stores information about the patient animal (hereinafter referred to as animal information). The animal information includes the name, animal species, and breed of the patient animal in association with the owner ID indicating the owner of the patient animal and the animal ID identifying the patient animal. In addition, animal information can include age, body weight, and prevention history. The age may be calculated by including the date of birth instead of the age. In addition, the body weight and prevention history (whether or not the patient was vaccinated or took medicine within a predetermined period) may be obtained from, for example, medical record information regarding medical treatment and examination of the patient animal. can. For example, the management server 20 can periodically access the medical record server that manages the medical record information, acquire the body weight and the prevention history, and update the animal information. The animal information can include various attributes of the patient animal.

The question storage unit 233 stores information related to a question about the health condition of the animal (hereinafter referred to as question information). The question information can include the question content in association with the question ID that identifies the question and the animal species to which the question corresponds. The content of the question includes a question that is scheduled to be answered in free text, a question that is scheduled to be answered by selecting one or more of two or three or more choices, and an answer with a value (real number or integer) within a predetermined range. Can include questions that are scheduled for. The question can be, for example, text data and can include images and sounds. In the present embodiment, the question includes a question to be answered by an image of a specific part.

The medical examination storage unit 234 stores the transmitted information regarding one medical examination (hereinafter referred to as medical examination information). The medical examination information includes the date and time when the medical examination was sent, the animal ID indicating the animal targeted for the medical examination, and the question ID indicating the question to be included in the medical examination, in association with the medical examination ID that identifies the medical examination. A plurality of question IDs may be set for one question ID.

The answer history storage unit 235 stores the history of answers from the owner to the question related to the medical examination. The answer history stored in the answer history storage unit 235 includes a question ID indicating a question and an answer to the question in association with the question ID, the animal ID, and the date and time when the question was conducted (or the answer was received). Is done. A set of question IDs and answers can be registered for one question ID as many as the number of questions.

The triage history storage unit 236 has a history of determining whether or not an animal patient needs to visit a veterinarian (assuming a veterinarian in this embodiment) with respect to the interview result (answer to the question). Triage history) is memorized. The diagnosis history includes the interview ID, the necessity of visiting the hospital, actions and comments. Whether or not a visit is necessary is either necessary or unnecessary. Actions are intervention information for animals, and are information indicating actions that animals should take, such as "let's take a walk" and "let's eat". Action options can be set in advance. The action may include the need for a visit. The comment is a comment on the result of the interview by the veterinarian. The comment can be, for example, text data. Comments may include possible illnesses.

The learning model storage unit 237 stores a learning model for determining the next question content (question items to be included in the question) based on the result of the question. The learning model storage unit 237 can store the learning model in association with the animal species. An animal species may be given as input data (features) of the learning model. Further, the learning model storage unit 237 can also store the learning model in association with the animal species and varieties. For example, even for the same dog, the questions to be asked may change between large dogs and small dogs. Therefore, it is also possible to store the learning model in association with the rank of body weight (the range determined by the maximum and minimum) in addition to or instead of the animal species and varieties. It is also possible to store the learning model in association with the age (range of age) in addition to or in place of at least one of the animal species and varieties and body weights. Further, the animal species and the breed may be given as the input data (features) of the learning model. In addition, at least one of body weight, body weight rank, age, and age may be included as the feature amount given to the learning model. In addition, past interview results can be included as the features given to the learning model. The past interview results can be characterized by adding a time lag (time information indicating how many times the interview was made or how many days ago the interview was made from the present time). As the teacher data given to learning the learning model, the next question to be asked determined by the veterinary professional can be adopted. That is, it is possible to learn the content of the next medical examination to be asked according to the answers to the past medical examinations determined by the veterinary medical staff by machine learning and use it as a learning model. The learning model is created and updated by the learning processing unit 216, which will be described later.

The medical examination transmission unit 211 periodically transmits the medical examination to the owner terminal 10. The inquiry transmission unit 211 can create one inquiry by selecting one or a plurality of questions stored in the inquiry storage unit 233 corresponding to the animal species of the patient animal. The medical examination transmission unit 211 can randomly select, for example, a predetermined number of questions corresponding to the animal species of the patient animal that satisfy the presentation conditions. Further, the medical examination transmission unit 211 may select a question according to a predetermined rule set in advance, for example.

The answer receiving unit 212 receives the answer to each question of the interview from the owner terminal 10. The answer receiving unit 212 can accept the upload of the image when the question includes the question to be answered by the image of the specific part. The answer receiving unit 212 can register the received answer in the answer history storage unit 235.

The visit necessity acquisition unit 215 can send a message requesting the medical institution terminal 30 to determine whether the visit is necessary or not. The visit necessity determination unit 213 can include the interview result (answer to the question, that is, the selection result of the option, the image, etc.) in the message and send it to the medical institution terminal 30. The medical institution terminal 30 can output the interview result and receive input of a judgment as to whether or not a visit is necessary from a medical person such as a veterinarian. The medical institution terminal 30 can also input intervention information (action) and a comment for the animal according to the result of the medical examination.

The visit necessity acquisition unit 215 can create a triage history including the visit necessity acquired from the medical institution terminal 30 and the inquiry ID indicating the inquiry corresponding to the inquiry result, and register it in the triage history storage unit 236. can. Further, when the action and the comment are transmitted from the medical institution terminal 30, the action and the comment can be included in the triage history.

The visit necessity transmission unit 214 transmits the visit necessity determined based on the interview result to the owner terminal 10.

The learning processing unit 216 creates and updates a learning model for determining the content of the medical examination. For example, when the interview storage unit 234 stores an interview created by a veterinary medical worker, the learning processing unit 216 reads out the interview content (question information) corresponding to the animal species for each animal species. Machine learning is performed using the question ID corresponding to the question (animal ID is the same and the date and time is the oldest among the question information with the same animal ID and the date and time is the oldest) sent after the question as input data as teacher data. The learning model created or updated by learning can be registered in the learning model storage unit 237. Since the interview can include multiple questions, each of the questions included in the next interview of one interview can be used as teacher data for learning.

The learning processing unit 216 can update the learning model periodically or at a timing determined by a predetermined rule. In this case, the learning processing unit 216 can send the inquiry content and the answer to the medical institution terminal 30, and can accept the input of the inquiry content (question ID) to be asked next from the veterinary medical worker, and input the answer. It is possible to update the learning model by performing machine learning using the question ID input from the veterinary medical staff as data as the teacher data.

<Operation>
FIG. 4 is a diagram illustrating the operation of the periodical examination system of the present embodiment.

The management server 20 creates a medical examination (S301). The interview can be created by reading out a predetermined number of questions stored in the question storage unit 233 that correspond to the animal species of the patient animal and satisfy the provision conditions. The management server 20 transmits the created interview to the owner terminal 10 (S302).

An interview is displayed on the owner terminal 10. The interview may be conducted, for example, in a chat format or in a form on a Web page. Further, for a question that is required to be answered by an image, the camera (not shown) provided in the owner terminal 10 can be activated and the captured image of the camera can be acquired as an answer. The owner terminal 10 transmits the input answer to the management server 20.

The management server 20 can receive an answer from the owner terminal 10 (S303) and register the received answer in the answer history storage unit 235 (S304). The management server 20 sends a message in which an answer is set to the medical institution terminal 30 to inquire about the necessity of visiting the hospital (S305). The medical institution terminal 30 receives input from a medical personnel such as a veterinarian and sends it to the management server 20, and the management server 20 receives the visit necessity from the medical institution terminal 30 and determines the visit necessity. It is registered in the triage history storage unit 236 (S306), and the necessity of visiting the hospital is transmitted to the owner terminal 10 (S307).

The management server 20 periodically repeats the process from S301 until the period for transmitting the periodic interview is completed (S308: NO). Before proceeding to S301, the management server 20 can give the answer to the interview to the learning model corresponding to the animal species, determine the question to be transmitted to the next inquiry, and register it in the inquiry storage unit 234 (S309). .. It should be noted that the date and time may not be set for the medical examination information here, or the date and time when the next medical examination is scheduled to be sent may be set.

As described above, in the periodic examination system of the present embodiment, the veterinary medical institution can proactively send a medical inquiry to the breeder of the patient animal to confirm the health condition of the patient animal. .. In addition, the next medical examination content can be automatically determined based on the medical examination result using AI (learning model).

Although the present embodiment has been described above, the above embodiment is for facilitating the understanding of the present invention, and is not for limiting the interpretation of the present invention. The present invention can be modified and improved without departing from the spirit thereof, and the present invention also includes an equivalent thereof.

For example, in the present embodiment, the learning model is created for each animal species, but one or a plurality of learning models may be used as a whole to give the animal species as a feature amount.

Further, in the learning model, the image taken by the owner terminal 10 can be included in the feature amount of the learning model. Thereby, for example, the content of the question to be asked next can be determined based on the photograph or the moving image of the onset site of the skin disease or the like, and the necessary observation according to the condition of the animal can be performed.

In addition, the necessity of visiting the hospital may be given as the input data (feature amount) of the learning model.

Further, the test result in the veterinary medical institution may be given as the input data (feature amount) of the learning model. In this case, for example, the management server 20 may manage the medical record information, or the medical record server that manages the medical record information may be accessed to acquire the inspection result included in the medical record information. can. In this case as well, the time lag can be used as input data together with the inspection result.

Further, in the present embodiment, the learning model is a classifier that determines the next question to be sent from a given question, but the present invention is not limited to this, and the generator that generates the next question to be asked is not limited to this. Or a transformer.

Further, in the present embodiment, the necessity of visiting the hospital is inquired to the veterinary medical staff, but the present invention is not limited to this, and for example, the necessity of visiting the hospital may be determined by AI. In this case, the classifier that determines the necessity of visiting the hospital can be stored in the learning model storage unit for each animal species. Alternatively, a classifier using the animal species as input data (feature amount) may be stored. The input data (features) of the classifier that requires a visit can be used as an answer to the interview.

10 Owner terminal 20 Management server 30 Medical institution terminal

Claims (8)

A system that manages the health of animals
The answer to the first question related to the health condition of the animal was used as input data, and the second question to be asked after the first question specified by the veterinary medical worker was created by machine learning as teacher data. A learning model storage unit that stores the learning model,
The medical examination transmission unit that sends the first question to the owner's terminal,
An answer receiving unit that receives the answer to the first question from the owner terminal, and an answer receiving unit.
An inference unit that gives the answer to the learning model and asks for the second question,
An animal health management system characterized by being equipped with. The animal health management system according to claim 1.
The learning model storage unit stores the learning model for each animal species, and stores the learning model.
The reasoning unit gives the answer to the learning model corresponding to the animal species of the animal.
An animal health management system featuring. The animal health management system according to claim 2.
The learning model storage unit stores the learning model in association with at least one of the animal species, breed, weight, body weight rank, age and age.
The reasoning unit provides the answer to the learning model corresponding to the animal species of the animal and at least one of the breed, the weight, the weight rank, the age and the age of the animal.
An animal health management system featuring. The animal health management system according to claim 1.
The learning model is learned by using the answer and the animal species as the input data.
The reasoning unit provides the learning model with the animal species of the animal and the answer.
An animal health management system featuring. The animal health management system according to claim 4.
The learning model is obtained by learning the answer, the animal species, and at least one of a breed, a body weight, a body weight rank, an age, and an age as the input data.
The reasoning unit gives the learning model the animal species of the animal, the answer, and at least one of the breed, the weight, the weight rank, the age, and the age of the animal.
An animal health management system featuring. The animal health management system according to any one of claims 1 to 5.
The learning model is learned by further adding the health condition of the animal determined by the veterinary medical worker to the input data.
Further provided with a health condition acquisition unit that outputs the answer to the veterinary medical worker and accepts the input of the health condition from the veterinary medical worker.
The reasoning unit further gives the input health state to the learning model.
An animal health management system featuring. The animal health management system according to any one of claims 1 to 6.
The learning model was learned by further adding the test results for the animal to the input data.
Further provided with an inspection result acquisition unit for acquiring the inspection result for the animal,
The inference unit further gives the acquired test result to the learning model.
An animal health management system featuring. The animal health management system according to any one of claims 1 to 7.
Further equipped with an answer history storage unit for storing the answer received from the owner terminal,
The learning model is learned by including the answer including the time lag in the input data.
The reasoning unit gives the learning model the answer stored in the answer history storage unit and the time lag from the present time.
An animal health management system featuring.

Images