Service providing method, service providing apparatus, and computer program

Abstract

To provide a service providing method, a service providing device, and a computer program.SOLUTION: A method for providing services to a plurality of users including an animal owner, a veterinarian of a primary medical institution, and a specialist of a secondary medical institution through a platform for linking a plurality of types of services related to animal care, the method comprising: A plurality of case data including animal observation data, diagnosis data of a veterinarian who has diagnosed the animal, and at least one of opinion data of a specialist on the observation data is collected, diagnosis support information for the veterinarian is generated based on the collected case data, the generated diagnosis support information is transmitted to a terminal device used by the veterinarian, and information on a supplement recommended to the animal diagnosed by the veterinarian is transmitted to a terminal device used by an owner.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to a service providing method, a service providing device, and a computer program. [Background technology]

[0002] Patent Document 1 discloses a medical support device that outputs support information for owners when acquiring information on the condition of their pets, and acquires the information on the condition of their pets based on the support information. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2020-140537 Summary of the Invention [Problem to be solved by the invention]

[0004] Currently, there is no platform that connects multiple types of veterinary medical services.

[0005] The present disclosure provides a service providing method, a service providing device, and a computer program that can provide services to users via a platform that links multiple types of veterinary care services. [Means for solving the problem]

[0006] The service provision method disclosed herein is a method of providing services to multiple users, including animal owners, veterinarians at primary medical institutions, and specialists at secondary medical institutions, via a platform for linking multiple types of veterinary medical services, in which a computer system executes the following processes: collects multiple case data including observation data of the animal, diagnostic data of the veterinarian who diagnosed the animal, and / or data on the specialist's findings on the observation data; generates diagnostic support information for the veterinarian based on the collected case data; transmits the generated diagnostic support information to a terminal device used by the veterinarian; and transmits information on supplements recommended for the animal diagnosed by the veterinarian to a terminal device used by the animal owner. [Effects of the Invention]

[0007] In the present disclosure, services can be provided to users via a platform that links multiple types of veterinary medical services. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is a schematic diagram illustrating a configuration example of a service providing system according to an embodiment. [Figure 2] FIG. 2 is a block diagram showing the internal configuration of the server device. [Figure 3] FIG. 2 is a conceptual diagram illustrating an example of the configuration of a management database. [Figure 4] FIG. 10 is a conceptual diagram showing an example of the configuration of a supplement database. [Figure 5] FIG. 2 is a schematic diagram showing an example of the configuration of a health state estimation model. [Figure 6] 10 is a flowchart illustrating a procedure of a process executed by a server device when providing a remote diagnosis service for pet owners. [Figure 7] FIG. 1 is an explanatory diagram illustrating an overview of an online medical consultation service for pet owners. [Figure 8] 10 is a flowchart illustrating a procedure of a process executed by a server device when providing an online medical consultation service for pet owners. [Figure 9]FIG. 10 is a schematic diagram showing an example of the configuration of a finding estimation model. [Figure 10] 10 is a flowchart illustrating a procedure of a process executed by a server device when providing a diagnostic support service for veterinarians. [Figure 11] FIG. 1 is an explanatory diagram illustrating an overview of an online medical consultation service for veterinarians. [Figure 12] 10 is a flowchart illustrating a procedure of a process executed by a server device when providing an online medical consultation service for veterinarians. [Figure 13] FIG. 10 is a schematic diagram illustrating an example of a sales screen. [Figure 14] FIG. 10 is a schematic diagram illustrating an example of a purchase contract screen. [Figure 15] 10 is a flowchart illustrating the processing steps executed by the server device when providing a supplement sales service. [Figure 16] FIG. 1 is an explanatory diagram illustrating an overview of a personalized supplement provision service. [Figure 17] 10 is a flowchart illustrating the processing steps executed by a server device when providing a personalized supplement provision service. DETAILED DESCRIPTION OF THE INVENTION

[0009] The present invention will now be described in detail with reference to the drawings showing embodiments thereof. (Embodiment 1) 1 is a schematic diagram showing an example of the configuration of a service providing system according to an embodiment. The service providing system 1 is a system for linking multiple types of services in a veterinary medical data platform business, an online medical consultation business, a diagnostic support business, and a supplement business via a platform 10, and providing services to multiple users, including animal owners, veterinarians at primary medical institutions, and veterinarians at secondary medical institutions.

[0010] The service providing system 1 includes a server device 100. The server device 100 is connected to a communication network NW such as the Internet. To the server device 100, an owner terminal 210 used by an animal owner, a veterinarian terminal 220 used by a veterinarian at a primary medical institution, a specialist terminal 230 used by a veterinarian at a secondary medical institution, an LLM (Large Language Model) server 300, and the like are communicatively connected via the communication network NW. The above-mentioned platform 10 is realized by the server device 100.

[0011] In an embodiment, the owner is any user who keeps an animal, such as a pet or livestock, that is the subject of a veterinarian's diagnosis. The animals that are the subject of a veterinarian's diagnosis are non-human animals such as dogs, cats, cows, horses, sheep, goats, pigs, and chickens. The animals that are the subject of a veterinarian's diagnosis may include stray dogs, stray cats, etc., regardless of whether they are kept by humans.

[0012] A veterinarian at a primary medical institution is a veterinarian who belongs to a local animal hospital or the like and diagnoses and treats animals at the request of owners. A veterinarian at a secondary medical institution is a veterinarian who belongs to a university hospital or the like and provides more specialized advice or diagnoses or treats animals at the request of, for example, a veterinarian at a primary medical institution. In this embodiment, to distinguish between veterinarians at a primary medical institution and veterinarians at a secondary medical institution, the former will be simply referred to as a veterinarian, and the latter will be referred to as a specialist. Specialists include internal medicine specialists, surgical specialists, pathologists, clinical pharmacologists, radiologists, etc. In this embodiment, it does not matter whether a specialist has been certified as a specialist by a specific organization.

[0013] The data platform business collects animal observation data, veterinarian diagnostic data, specialist findings based on the observation data, literature data on animal diagnoses, and data on supplements recommended by veterinarians or specialists, and provides an environment for utilizing the collected data. For example, the animal observation data and veterinarian diagnostic data collected by the data platform business are provided to the online medical consultation business to be used as training data for generating a learning model (learning model LM1, see Figure 5) that estimates the health status of animals from the observation data. The animal observation data and specialist findings collected by the data platform business are provided to the diagnostic support business to be used as training data for generating a learning model (learning model LM2, see Figure 9) that estimates the specialist findings from the observation data. Furthermore, the veterinarian diagnostic data and diagnostic literature data collected by the data platform business may be provided to the online medical consultation business and the diagnostic support business to be used as reference data for language models. Furthermore, supplement data may be provided to supplement manufacturers and distributors.

[0014] The online medical consultation business provides online services for pet owners regarding animal diagnoses or medical consultations. For example, the online medical consultation business provides a remote diagnosis service in which observation data of an animal to be diagnosed is received from an owner terminal 210 via a communication network NW, and an estimated result regarding the animal's health condition obtained by inputting the received observation data into a learning model LM1 is returned to the owner terminal 210 via the communication network NW. Note that remote diagnosis is a term that refers to a diagnosis made at a location that is too far away to receive a face-to-face diagnosis from a veterinarian or specialist, regardless of whether the location is physically far away.

[0015] The online medical care business may provide a service in which, instead of the estimation results by the learning model LM1, a diagnosis result made by a veterinarian is returned to the owner via the communication network NW. The online medical care business may also provide an online medical consultation service in which an owner makes a medical consultation about an animal and returns response data to the medical consultation.

[0016] The diagnostic support business provides online services for veterinarians regarding diagnostic support or medical consultation. For example, the diagnostic support business receives observation data of an animal to be diagnosed from a veterinarian terminal 220 via a communication network NW, and returns estimated findings obtained by inputting the received observation data into a learning model LM2 to the veterinarian terminal 220 via the communication network NW.

[0017] The diagnostic business may provide a service that returns diagnostic results from a specialist to a veterinarian via a communication network NW instead of the estimation results from the learning model LM2. The diagnostic support business may also provide an online medical consultation service that accepts medical consultations about animals from veterinarians and returns response data to those consultations.

[0018] The supplement business provides online services that provide information on supplements recommended by veterinarians or specialists to pet owners and provide information on supplements purchased by pet owners to manufacturers and distributors. For example, the supplement business provides an online sales service that displays a purchase contract screen including supplements recommended by veterinarians or specialists on the pet owner's terminal 210 and accepts purchase contracts through the purchase contract screen. The supplement business may also provide an information provision service that provides information on supplements purchased by pet owners, information on the pet owners and the animals they keep, etc. to manufacturers and distributors of supplements.

[0019] In the example of Fig. 1, the platform 10 is configured to be realized by a single server device 100. Alternatively, the platform 10 may be realized by a computer system including client devices, server devices, storage devices, and other peripheral devices used by each business operator. Also, in the example of Fig. 1, the number of each of the owner terminal 210, veterinarian terminal 220, and specialist terminal 230 connected to the communication network NW is one, but multiple terminal devices may be connected to each.

[0020] 2 is a block diagram showing the internal configuration of the server device 100. The server device 100 is a dedicated or general-purpose server computer, and includes a control unit 101, a storage unit 102, a communication unit 103, an operation unit 104, a display unit 105, and the like.

[0021] The control unit 101 is a processing circuit or arithmetic circuit including, for example, a CPU (Central Processing Unit), a ROM (Read Only Memory), and a RAM (Random Access Memory). The CPU included in the control unit 101 reads and executes various computer programs stored in the ROM or the storage unit 102, thereby controlling each hardware unit and causing the entire device to function as the service providing device of the present disclosure. The RAM included in the control unit 101 temporarily stores data used during the execution of calculations.

[0022] In the embodiment, the control unit 101 is configured to include a CPU, a ROM, and a RAM, but the configuration of the control unit 101 is not limited to the above. The control unit 101 may be, for example, one or more control circuits or processing circuits including a GPU (Graphics Processing Unit), an FPGA (Field Programmable Gate Array), a DSP (Digital Signal Processor), a quantum processor, volatile or non-volatile memory, etc. The control unit 101 may also have functions such as a clock that outputs date and time information, a timer that measures the elapsed time from when a measurement start instruction is given until when a measurement end instruction is given, and a counter that counts numbers.

[0023] The storage unit 102 includes a storage device such as a hard disk, a flash memory, etc. The storage unit 102 stores various computer programs executed by the control unit 101 and various data used by the control unit 101. The computer programs stored in the storage unit 102 include a service providing program PG that causes a computer to execute the following processes: collect a plurality of case data including animal observation data and at least one of veterinarian diagnosis data and specialist finding data; generate diagnostic support information for veterinarians based on the collected case data; transmit the generated diagnostic support information to the veterinarian terminal 220; and transmit information on supplements recommended for the animal diagnosed by the veterinarian to the owner terminal 210.

[0024] The service providing program PG may be a single computer program or a group of programs consisting of multiple computer programs. The service providing program PG may be executed by multiple computers working together. The service providing program PG may partially use an existing library.

[0025] A computer program (program product) including the service provision program PG is provided, for example, by a non-transitory recording medium RM on which the computer program is readably recorded. The recording medium RM is a portable memory such as a CD-ROM, a USB memory, an SD (Secure Digital) card, a micro SD card, or a CompactFlash (registered trademark). The control unit 101 reads various computer programs from the recording medium RM using a reading device (not shown) and stores the read various computer programs in the storage unit 102. The computer programs stored in the storage unit 102 may also be provided via communication. In this case, the control unit 101 acquires the computer programs by communication via the communication unit 103 and stores the acquired computer programs in the storage unit 102.

[0026] The storage unit 102 may also include a management database DB1 that stores information about animals examined by veterinarians or specialists, a supplement database DB2 that stores information about supplements for animals, etc. The contents stored in each database DB1 and DB2 will be described with reference to FIGS.

[0027] The communication unit 103 includes a communication interface for transmitting and receiving various data to and from an external device. A wired or wireless communication interface conforming to a communication standard such as a LAN (Local Area Network) can be used as the communication interface of the communication unit 103. The external device may be an owner terminal 210, a veterinarian terminal 220, or a specialist terminal 230. When data to be transmitted is input from the control unit 101, the communication unit 103 transmits the data to the destination external device, and when data transmitted from an external device is received, the communication unit 103 outputs the received data to the control unit 101.

[0028] The operation unit 104 includes operation devices such as a touch panel, a keyboard, and switches, and receives various inputs and operations from a user, etc. The control unit 101 acquires information input through the operation unit 104 and performs appropriate control based on various operation information provided by the operation unit 104.

[0029] The display unit 105 includes a display device such as a liquid crystal monitor or an organic EL (Electro-Luminescence) monitor, and displays information to be notified to the user or the like in response to an instruction from the control unit 101.

[0030] In this embodiment, the server device 100 is a single server computer, but the server device 100 may be a computer system configured with multiple computers, peripheral devices, etc. Furthermore, the server device 100 may be a virtual machine whose entity is virtualized, or may be a cloud.

[0031] In this embodiment, the server device 100 is configured to include the management database DB1 and the supplement database DB2, but these databases DB1 and DB2 may be provided outside the device as long as they are accessible from the server device 100.

[0032] FIG. 3 is a conceptual diagram showing an example of the configuration of the management database DB1. The management database DB1 stores, in association with each other, the date and time, the sender, the recipient, the target animal, observation data, diagnostic data, and finding data. The date and time represent the date and time when the recipient received the data sent from the sender. The sender and the recipient represent identifiers that identify the owner, veterinarian, and specialist. Information such as name, address, telephone number, and email address may be stored in association with each identifier. The target animal represents an identifier that identifies the animal. Information such as the animal's type, sex, age, weight, and image may be stored in association with the animal's identifier.

[0033] The observation data includes at least one of image data, vital sign data, audio data related to animal cries, olfactory data, and data on the owner's responses to medical interviews. The observation data is data obtained by a veterinarian or specialist observing an animal using medical equipment. The observation data may also be data obtained by the owner observing the animal using a smartphone or the like.

[0034] Of the observation data, image data is image data captured as medical images in tests such as MRI (Magnetic Resonance Imaging), CT (Computed Tomography), and endoscopic examinations. The image data may be image data obtained by using a general optical camera to capture images of an animal's appearance, affected areas, excrement, etc. Vital data is data such as an animal's blood pressure, pulse, sweat, and activity level measured using existing measuring instruments or sensors. Audio data is digital audio data obtained by recording the animal's cries. Olfactory data is data obtained by measuring (quantifying) the odor of an animal or its excrement using an olfactory sensor. Response data is data of the owner's responses to medical questions such as "Is the animal eating?", "Is it active?", "Is it calm?", and "Are there any abrasions?"

[0035] The diagnosis data represents the diagnosis results by a veterinarian and includes text data, image data, etc. The findings data represents the findings of a specialist on the animal observation data and includes text data, image data, etc.

[0036] Whenever the server device 100 acquires observation data, diagnosis data, or finding data from the owner terminal 210, the veterinarian terminal 220, or the specialist terminal 230, it registers the acquired data in the management database DB1.

[0037] Figure 4 is a conceptual diagram showing an example of the configuration of the supplement database DB2. The supplement database DB2 stores information such as the name of the supplement, cases, manufacturer, seller, price, and recommended doctor in association with each other. The name represents the product name of the supplement. The case represents the case of an animal for which the supplement is effective. The manufacturer and seller represent the names of the manufacturer and seller of the supplement, respectively. The price represents the selling price of the supplement. The price may be the price per a specified period (e.g., per month). The recommended doctor represents the identifier of the veterinarian or specialist who recommends the supplement.

[0038] When the server device 100 receives data on supplements from a seller or the like, it registers the received data on the supplements in the supplement database DB2 each time.

[0039] The server device 100 utilizes data stored in the management database DB1, supplement database DB2, etc. to provide services to multiple users (pet owners, veterinarians, specialists, etc.). The services provided by the server device 100 include (1) remote diagnosis for pet owners, (2) online medical consultation for pet owners, (3) diagnostic support for veterinarians, (4) online medical consultation for veterinarians, and (5) sales of supplements to pet owners. Each of these services will be explained below.

[0040] (1) Remote diagnosis for pet owners As the collection of animal observation data and veterinarian diagnosis data progresses in the server device 100, machine learning can be performed using a dataset including the collected observation data and diagnosis data as training data to generate a first learning model LM1 (hereinafter referred to as health condition estimation model LM1) that estimates the health condition of an animal from the animal observation data. By using this health condition estimation model LM1, the server device 100 provides a remote diagnosis service for owners.

[0041] Figure 5 is a schematic diagram showing an example configuration of health condition estimation model LM1. The health condition estimation model LM1 in Figure 5 is a learning model generated by machine learning including deep learning, and is composed of a neural network including a first network layer NL11, a second network layer NL12, and a third network layer NL13.

[0042] For example, when image data, which is one of the observation data, is input, the first network layer NL11 is configured to extract features from the input image data and output the extracted features to the subsequent third network layer NL13. For example, a convolutional neural network (CNN) including a convolution layer and a pooling layer is used for this first network layer NL11.

[0043] For example, when audio data, which is one of the observation data, is input, the second network layer NL12 is configured to extract features from the input audio data and output the extracted features to the subsequent third network layer NL13. For example, an autoencoder equipped with an encoder and a decoder is used in this second network layer NL12, and the features extracted by the encoder are output to the subsequent third network layer NL13.

[0044] The third network layer NL13 includes, for example, a concatenation layer, a smoothing layer, a linear layer, and a dropout layer, and is configured to concatenate the features of the image data output from the first network layer NL11 and the features of the audio data output from the second network layer NL12 in the concatenation layer, and to output information about the animal's health by sequentially processing the output of the concatenation layer in each subsequent layer. The final linear layer is constructed using a softmax layer, and outputs information about the animal's health, such as the probability P1 that the animal is estimated to be healthy (or the probability P2 that the animal is suffering from a disease).

[0045] Server device 100 can generate a health state estimation model LM1 by learning using an existing learning algorithm. For example, server device 100 reads observation data stored in management database DB1 and diagnostic data corresponding to the observation data, and performs supervised learning using the veterinarian's diagnosis results included in the diagnostic data as ground truth data to generate a health state estimation model LM1 configured to output an estimation result related to an animal's health state when observation data is input. The generated health state estimation model LM1 may be stored in storage unit 102 or in an external storage device accessible from server device 100.

[0046] In the example of Figure 5, image data and audio data from observation data are input into the health condition estimation model LM1 to estimate the health condition of an animal. However, the observation data input into the health condition estimation model LM1 is not limited to image data and audio data, and may be any one of image data, vital data, audio data, olfactory data, and response data, or a combination of any two or more of these. Furthermore, the configuration of the network layer and each network layer constituting the health condition estimation model LM1 is not limited to the CNN or autoencoder illustrated in Figure 5, and may be appropriately designed depending on the observation data to be input. Furthermore, the server device 100 may estimate the health condition of an animal using a multimodal language model.

[0047] In addition, in the example of Figure 5, the configuration is such that the probability P1 that the target animal is estimated to be healthy (or the probability P2 that the animal is suffering from a disease) is output, but the health status estimation model LM1 may also be trained to estimate the type of disease the animal is suffering from and output the estimation result, or may be trained to output the probability of suffering from each of multiple types of disease.

[0048] Server device 100 provides a remote diagnosis service to owners using the trained health state estimation model LM1. FIG. 6 is a flowchart illustrating the steps of the process executed by server device 100 when providing the remote diagnosis service for owners. It is assumed that server device 100 stores trained health state estimation model LM1 in memory unit 102 of server device 100. Server device 100 accepts a diagnosis request from owner terminal 210 via communication unit 103, and receives observation data of the animal to be diagnosed from owner terminal 210 (step S101). After accepting the diagnosis request from owner terminal 210, server device 100 may request the observation data and receive the observation data returned from owner terminal 210 in response to the request, or may receive the observation data sent from owner terminal 210 together with the diagnosis request.

[0049] The control unit 101 of the server device 100 inputs the received observation data into the health state estimation model LM1 and executes calculations using the health state estimation model LM1 (step S102). The control unit 101 estimates the health state of the animal based on the calculation results using the health state estimation model LM1 (step S103). If the health state estimation model LM1 has been trained to output a probability P1 that the target animal will be estimated to be healthy, the control unit 101 compares the probability P1 with a preset threshold (e.g., 80%), and if the probability P1 is equal to or greater than the threshold, it estimates that the animal is healthy, and if the probability P1 is less than the threshold, it estimates that the animal may not be healthy. The threshold is not limited to 80% and can be set as appropriate.

[0050] The control unit 101 transmits the estimation result of step S103 (the estimation result related to the health condition of the animal kept by the owner) to the owner terminal 210 via the communication unit 103 (step S104).

[0051] The owner terminal 210 displays the estimated results received from the server device 100 on the display unit of the terminal. The owner can obtain the estimated results regarding the health condition of the animal remotely without having to take the animal to a primary or secondary medical institution for a face-to-face diagnosis.

[0052] In this embodiment, the estimation results using the health state estimation model LM1 are provided to the owner. However, the platform 10 may relay communication between the owner terminal 210 and the veterinarian terminal 220 to provide a remote diagnosis support service to the owner. In this case, observation data of the animal is transmitted from the owner terminal 210 to the veterinarian terminal 220, and the veterinarian's diagnosis results based on the observation data output from the veterinarian terminal 220 are returned from the veterinarian terminal 220 to the owner terminal 210. At this time, information about supplements recommended by the veterinarian may be transmitted from the veterinarian terminal 220 to the owner terminal 210. Furthermore, a supplement purchase contract screen (see FIG. 14, for example) may be displayed on the display unit of the owner terminal 210 so that the owner can purchase the supplements recommended by the veterinarian.

[0053] Server device 100 may also input the animal observation data received from owner terminal 210 into health state estimation model LM1 to estimate the animal's health state and provide the estimation result to the primary medical institution. In this case, the primary medical institution makes a diagnosis by referring to the estimation result of the animal's health state provided by server device 100, and transmits the diagnosis result from veterinarian terminal 220 to owner terminal 210.

[0054] Furthermore, the observation data transmitted by the owner is not limited to observation data obtained by observing using a smartphone. For example, the owner may observe the animal using a dedicated or general-purpose measuring device that measures the animal's blood pressure, pulse rate, sweating, activity level, vocalizations, odor, etc., and transmit the obtained observation data to the server device 100 via the owner terminal 210. Furthermore, the measuring device that measures the above observation data may be sold or rented to the owner by a measuring device manufacturer or medical device manufacturer that cooperates with the business operator through the platform 10.

[0055] (2) Online medical consultation for pet owners The server device 100 collects case data including animal observation data and at least one of veterinarian diagnosis data and specialist findings data. When the server device 100 receives speech data including the content of a consultation from a pet owner, it refers to the collected case data, generates response data to the received speech data, and returns the generated response data, thereby providing an online medical consultation service for pet owners.

[0056] FIG. 7 is an explanatory diagram illustrating an overview of an online medical consultation service for pet owners. The server device 100 receives speech data including a consultation content (query) from an animal owner via a dedicated app such as a chat app or a web browser. The speech data may be text data or voice data. When receiving the speech data in the form of voice data, the server device 100 may convert the voice data into text data using an existing conversion algorithm.

[0057] The server device 100 generates input data (also called a prompt) for the language model 310 based on speech data including the content of the owner's consultation. For example, the server device 100 generates a prompt specifying that the role is a veterinarian, that the management database DB1 should be referenced as reference information, and that an answer should be given to the content of the owner's consultation. In addition to the management database DB1, the server device 100 may also specify the supplement database DB2 as reference information, or may specify literature data related to animal diagnosis available via the communication network NW as reference information.

[0058] The server device 100 transmits the generated prompt to the LLM server 300. The LLM server 300 includes a language model 310, and generates a response sentence by inputting the prompt received from the server device 100 into the language model 310. The language model 310 is an existing large-scale language model such as GPT-4 (Generative Pretrained Transformer 4), LLaMA (Large Language Model Meta AI), or BERT (Bidirectional Encoder Representations from Transformers). The language model 310 is not limited to the above language models, and may be a unique language model tuned from an existing language model, or may be a multimodal language model that can also input image data, audio data, and the like. In the latter case, animal observation data may be input to the language model 310 in addition to the text-based prompt.

[0059] By inputting the above prompt into the language model 310, the LLM server 300 generates a response based on past case data, including the veterinarian's diagnosis or the specialist's findings, in response to the owner's inquiry. The response includes information on how to deal with the inquiry, such as first aid, whether or not to visit a veterinary clinic, and recommended supplements. The LLM server 300 returns the generated response to the server device 100.

[0060] The server device 100 generates answer data based on the answer text received from the LLM server 300 and transmits the generated answer data to the owner terminal 210. The owner terminal 210 displays the answer data received from the server device 100 on a display unit. The owner can obtain answers to the consultation content, such as how to deal with the issue based on past case data.

[0061] 8 is a flowchart illustrating the processing steps executed by the server device 100 when providing an online medical consultation service for pet owners. When a pet owner wishes to have an online medical consultation, the pet owner launches a dedicated app or a web browser on the pet owner terminal 210 and accesses the server device 100. When speech data including the content of the pet owner's consultation is input, the pet owner terminal 210 transmits the input speech data to the server device 100.

[0062] The server device 100 receives speech data including the owner's consultation content from the owner terminal 210 via the communication unit 103 (step S201). The control unit 101 of the server device 100 extracts the owner's consultation content (query) from the received speech data, and generates a prompt that includes the extracted consultation content and specifies a response from the veterinarian by referring to the contents of the management database DB1 (and literature data on the communication network NW) (step S202). It is assumed that a template for the prompt is prepared in advance and stored in the memory unit 102. The control unit 101 reads the template from the memory unit 102 and generates the prompt by embedding the owner's consultation content in the read template.

[0063] The control unit 101 transmits the generated prompt to the LLM server 300 (step S203). The LLM server 300 receives the prompt, inputs the received prompt into the language model 310, generates a response sentence according to the content of the owner's consultation, and returns the generated response sentence to the server device 100.

[0064] The control unit 101 receives the reply message returned from the LLM server 300 via the communication unit 103 (step S204), and generates reply data for the owner based on the received reply message (step S205). The control unit 101 transmits the reply data generated in step S205 to the owner terminal 210 (step S206).

[0065] The owner terminal 210 displays the response data received from the server device 100 on its display unit. The owner can ask the server device 100 about the content of their inquiry whenever they think of it, regardless of time or place, and can receive a response such as a solution based on past case data.

[0066] (3) Diagnostic support for veterinarians As the collection of animal observation data and specialist finding data progresses in the server device 100, machine learning is performed using a dataset including the collected observation data and finding data (e.g., radiology reports) as training data, and a learning model LM2 (hereinafter referred to as finding estimation model LM2) that estimates the specialist finding from the animal observation data can be generated. By using this finding estimation model LM2, the server device 100 provides a diagnostic support service for veterinarians.

[0067] Fig. 9 is a schematic diagram showing an example configuration of the finding estimation model LM2. The finding estimation model LM2 in Fig. 9 is, for example, a learning model generated by machine learning including deep learning, and is configured by a CNN having an input layer NL21, an intermediate layer NL22, and an output layer NL23. Alternatively, the finding estimation model LM2 may be constructed using a neural network for object detection such as YOLO (You Only Look Once) or SSD (Single Shot Multi-Box Detector), or may be constructed using a neural network for image segmentation such as SegNet, FCN (Fully Convolutional Network), U-Net (U-Shaped Network), or PSPNet (Pyramid Scene Parsing Network).

[0068] Observation data is input to the input layer NL21 of the finding estimation model LM2. The observation data input to the input layer NL21 is data that requires a specialist's opinion. For example, the observation data is image data related to medical images obtained by MRI examinations, CT examinations, endoscopic examinations, etc. The middle layer NL22 includes a convolutional layer, a pooling layer, a fully connected layer, etc. The calculations performed by the middle layer NL22 include processing to extract features from the image data, processing to select candidate regions likely to be the detection target contained in the image data, and processing to identify the detection target within the selected region. An example of a detection target is a lesion. The output layer NL23 references the calculation results from the middle layer and outputs the estimation results from the finding estimation model LM2. The estimation results include, for example, information about the lesion contained in the medical image (the type of lesion, the location of the lesion within the medical image).

[0069] The server device 100 can generate the finding estimation model LM2 by learning using an existing learning algorithm. For example, the server device 100 reads out observation data stored in the management database DB1 and the finding data corresponding to the observation data, and performs supervised learning using the specialist's findings included in the finding data as correct answer data to generate the finding estimation model LM2 configured to estimate the specialist's findings when the observation data is input. The generated finding estimation model LM2 may be stored in the storage unit 102 or in an external storage device accessible from the server device 100.

[0070] 9 shows a finding estimation model LM2 that estimates a specialist's findings when image data related to a medical image is input, but the finding estimation model LM2 is not limited to the above and may be a unimodal learning model or multimodal learning model that estimates a specialist's findings when any one or any combination of two or more of image data showing an animal's appearance, vital data, audio data, olfactory data, and response data is input. The configuration of the network layers and each network layer that make up the finding estimation model LM2 is not limited to the CNN shown in FIG. 9, and is designed appropriately depending on the observation data to be input.

[0071] The server device 100 provides a diagnostic support service to veterinarians using the trained finding estimation model LM2. FIG. 10 is a flowchart illustrating the steps of a process executed by the server device 100 when providing a diagnostic support service for veterinarians. The server device 100 is assumed to store the trained finding estimation model LM2 in the memory unit 102 of the server device 100. The server device 100 accepts a diagnosis request from the veterinarian terminal 220 via the communication unit 103, and receives observation data of the animal to be diagnosed (e.g., image data related to medical images) from the specialist terminal 230 (step S301). After accepting the diagnosis request from the veterinarian terminal 220, the server device 100 may request the observation data and receive the observation data returned from the veterinarian terminal 220 in response to the request, or may receive the observation data sent from the owner terminal 210 together with the diagnosis request.

[0072] The control unit 101 of the server device 100 inputs the received observation data into the finding estimation model LM2 and executes calculations using the finding estimation model LM2 (step S302). The control unit 101 estimates the specialist's findings based on the calculation results using the finding estimation model LM2 (step S303).

[0073] The control unit 101 transmits the estimation result of step S303 to the veterinarian terminal 220 via the communication unit 103 (step S304). The control unit 101 may generate information on at least one of a treatment method, a therapeutic drug, and a supplement for the animal as diagnostic support information based on the estimation result by the finding estimation model LM2, and transmit the generated information to the veterinarian terminal 220. The information on a treatment method, a therapeutic drug, and a supplement for the animal can be generated by using a language model 320 described below, or the like.

[0074] The veterinarian terminal 220 displays the inference results received from the server device 100 on the display unit of the terminal. The veterinarian can obtain the inference results related to the specialist's findings remotely without having to visit a secondary medical institution and receive a face-to-face diagnosis from the specialist.

[0075] (4) Medical consultation for veterinarians The server device 100 collects case data including animal observation data and at least one of veterinarian diagnosis data and specialist findings data. When the server device 100 receives speech data including the content of a consultation with a veterinarian, it refers to the collected case data, generates response data to the received speech data, and returns the generated response data to the specialist, thereby providing an online medical consultation service for specialists.

[0076] FIG. 11 is an explanatory diagram illustrating an overview of an online medical consultation service for veterinarians. The server device 100 receives speech data including a consultation content (query) from a veterinarian via a dedicated app such as a chat app or a web browser. The speech data may be text data or voice data. When receiving the speech data in the form of voice data, the server device 100 may convert the voice data into text data using an existing conversion algorithm.

[0077] The server device 100 generates input data (prompts) for the language model 310 based on speech data including the content of the consultation with the veterinarian. For example, the server device 100 generates a prompt specifying that the role is a specialist, that the management database DB1 should be referenced as reference information, and that the veterinarian should respond to the content of the consultation. In addition to the management database DB1, the server device 100 may also specify literature data related to animal diagnosis available via the communication network NW as reference information.

[0078] The server device 100 transmits the generated prompt to the LLM server 300. The LLM server 300 includes a language model 320, and generates a response sentence by inputting the prompt received from the server device 100 into the language model 320. The language model 320 may be the same as that shown in FIG. 7, but since observation data (image data related to medical images such as MRIs) is often presented in medical consultations for veterinarians, it is preferable that the language model 320 be a multimodal language model that can input image data, etc. Furthermore, the language model 320 may be a model fine-tuned for veterinarians.

[0079] The LLM server 300 generates a response based on past case data, including the findings of specialists, in response to the veterinarian's consultation by inputting the above-mentioned prompt into the language model 320. The LLM server 300 returns the generated response to the server device 100.

[0080] The server device 100 generates response data based on the response text received from the LLM server 300 and transmits the generated response data to the veterinarian terminal 220. The veterinarian terminal 220 displays the response data received from the server device 100 on a display unit. The veterinarian can obtain a response to the consultation content, such as a finding based on past case data.

[0081] 12 is a flowchart illustrating the steps of the process executed by the server device 100 when providing an online medical consultation service for veterinarians. When a veterinarian conducts an online medical consultation, the veterinarian launches a dedicated app or a web browser on the veterinarian terminal 220 and accesses the server device 100. When speech data including the content of the veterinarian's consultation is input, the veterinarian terminal 220 transmits the input speech data to the server device 100.

[0082] The server device 100 receives speech data including the veterinarian's consultation content from the veterinarian terminal 220 via the communication unit 103 (step S401). The control unit 101 of the server device 100 extracts the veterinarian's consultation content (query) from the received speech data, and generates a prompt that includes the extracted consultation content and specifies a response from the specialist by referring to the contents of the management database DB1 (and literature data on the communication network NW) (step S402). It is assumed that a template for the prompt is prepared in advance and stored in the memory unit 102. The control unit 101 reads the template from the memory unit 102 and generates the prompt by embedding the veterinarian's consultation content in the read template.

[0083] The control unit 101 transmits the generated prompt to the LLM server 300 (step S403). The LLM server 300 receives the prompt, inputs the received prompt into the language model 320, generates a response sentence according to the consultation content of the veterinarian, and returns the generated response sentence to the server device 100.

[0084] The control unit 101 receives the reply message sent back from the LLM server 300 via the communication unit 103 (step S404), and generates reply data for the veterinarian based on the received reply message (step S405). The control unit 101 transmits the reply data generated in step S405 to the veterinarian terminal 220 (step S406).

[0085] The veterinarian terminal 220 displays the response data received from the server device 100 on the display unit. The veterinarian can receive a response such as a finding based on past case data by transmitting the consultation details to the server device 100 regardless of time or place.

[0086] (5) Sales of supplements for pet owners The server device 100 provides a sales service for supplements to pet owners based on information registered in the supplement database DB2.

[0087] The server device 100 publishes the supplement sales website managed in the supplement database DB2 on the communication network NW. When the server device 100 receives access to the sales website from the owner terminal 210, it displays a supplement sales screen on the display unit of the owner terminal 210.

[0088] FIG. 13 is a schematic diagram showing an example of a sales screen. The sales screen 130 shown in FIG. 13 is an interface screen that allows the owner to operate and input information, and includes a product search field 131 for accepting product searches, a product display field 132, product selection buttons 133a, 133b, 133c, ..., a chat button 134, and a product purchase button 135. When keywords such as a product name or symptoms are entered in the product search field 131, the owner terminal 210 transmits the entered search keywords to the server device 100. When the server device 100 receives the search keywords from the owner terminal 210, it searches the supplement database DB2 and displays information about the supplements (products) obtained as a search result in the product display field 132. At this time, if the supplement is recommended by a veterinarian or specialist, this information is displayed in the product display field 132.

[0089] When the pet owner finds a supplement they wish to purchase, they select it by pressing product selection buttons 133a, 133b, 133c, ... and add the selected product to their cart. At this time, the pet owner may press chat button 134 to ask for medical advice regarding the purchase of the supplement. The pet owner conducts a text-based consultation on the chat screen that appears when chat button 134 is pressed. Server device 100 generates a response to the pet owner's inquiry using language model 310 that references management database DB1 and supplement database DB2, and returns the generated response to pet owner terminal 210. The specific processing procedure is the same as that described in the flowchart of FIG. 8.

[0090] When the product purchase button 135 is pressed on the sales screen 130, the server device 100 transitions the display on the owner terminal 210 to a purchase contract screen.

[0091] FIG. 14 is a schematic diagram showing an example of a purchase contract screen. The purchase contract screen 140 shown in FIG. 14 is an interface screen that allows the owner to operate and input information, and includes a product confirmation field 141, an application confirmation field 142, a payment method selection field 143, and an order confirmation button 144. The product confirmation field 141 displays information about the product that the owner added to the cart on the sales screen 130. The application confirmation field 142 displays the details of the purchase application. The payment method selection field 143 accepts the selection of a payment method and accepts input of necessary information depending on the payment method. The purchase contract screen 140 may also display other information such as the product shipping address, shipping method, shipping time, and contract cancellation method. When the order confirmation button 144 is pressed on the purchase contract screen 140, the control unit 101 of the server device 100 stores the contract details in the memory unit 102.

[0092] 15 is a flowchart illustrating the procedure of the process executed by the server device 100 when providing a supplement sales service. The control unit 101 of the server device 100 determines whether or not access to the sales site has been accepted from the owner terminal 210 (step S501). If access to the sales site has not been accepted (S501: NO), the control unit 101 waits until access to the sales site is accepted.

[0093] When access to the sales site is accepted (S501: YES), the control unit 101 displays the sales screen 130 shown in Fig. 13 on the display unit of the owner terminal 210 (step S502). When a search keyword is accepted on the sales screen 130, the control unit 101 searches the supplement database DB2 for products (supplements) that match the search keyword, and displays the information obtained as the search result in the product display field 132 on the sales screen 130.

[0094] The control unit 101 determines whether the owner has selected a product they wish to purchase on the sales screen 130 and pressed the product purchase button 135 (step S503). If the sales screen 130 is closed without selecting a product (S503: NO), the control unit 101 ends the processing according to this flowchart.

[0095] When a product is selected on the sales screen 130 and the product purchase button 135 is pressed (S503: YES), the control unit 101 displays a purchase contract screen 140 as shown in FIG. 14 on the display unit of the owner terminal 210 (step S504).

[0096] The control unit 101 determines whether or not information such as the payment method and the shipping address of the product has been entered on the purchase contract screen 140 and the order confirmation button 144 has been pressed (step S505). If the order confirmation button 144 has not been pressed (S505: NO), the control unit 101 waits until the order is confirmed.

[0097] When information such as the payment method and the shipping address of the product is entered and the order confirmation button 144 is pressed (S505: YES), the control unit 101 stores the confirmed contract details in the storage unit 102 (step S506).

[0098] The control unit 101 determines whether it is time to ship the product based on the contract details stored in the memory unit 102 (step S507), and if it is not time to ship (S507: NO), it waits, and if it is time to ship (S507: YES), it outputs a product shipping instruction to the supplement distributor (step S508).

[0099] As described above, in the embodiment, multiple types of veterinary medical services, including a remote diagnosis service for pet owners, an online medical consultation service for pet owners, a diagnostic support service for veterinarians, an online medical consultation service for veterinarians, and a supplement sales service for pet owners, can be linked and provided to users via platform 10.

[0100] (Embodiment 2) In the second embodiment, a form will be described in which supplements suited to individual animals (hereinafter referred to as individualized supplements) are provided to owners. The overall configuration of the service providing system 1 and the configuration of each individual device are the same as those in the first embodiment, and therefore the description thereof will be omitted.

[0101] FIG. 16 is an explanatory diagram outlining the personalized supplement provision service. A supplement business operator opens an online site on a communications network NW via the platform 10 to provide personalized supplement provision services to pet owners. The operator accepts account registrations from pet owners through the online site and concludes subscription contracts with pet owners who wish to purchase personalized supplements on a regular basis. Information on pet owners who have concluded subscription contracts and information on the animals they keep is stored, for example, in the memory unit 102 of the server device 100. A secondary medical institution such as a university hospital obtains information on pet owners and animals through the platform 10 and develops personalized supplements in collaboration with the operator. Information on the jointly developed personalized supplements is provided to pet owners as information on recommended supplements.

[0102] The personalized supplement may be provided to the pet owner based on the test results of a blood test, a urine test, a genetic test, an imaging test, etc. For example, when a blood test using an animal's blood as a sample is conducted at a secondary medical institution, the test results are notified to the pet owner terminal 210 via the platform 10, and personalized supplements (e.g., supplements containing essential amino acids, etc.) based on the test results may be provided to the pet owner by a supplement manufacturer / distributor upon request from the platform 10. The timing and frequency of providing the personalized supplement are determined in advance by the subscription contract.

[0103] The same applies to urine tests, genetic tests, imaging tests, etc. In urine tests, a renal function test is conducted using the animal's urine as a sample, and depending on the test results, a personalized supplement (e.g., a supplement containing high protein, vitamins, minerals, etc.) that improves renal function may be provided to the owner. In genetic tests, an intestinal environment test is conducted using the animal's feces as a sample, and depending on the test results, a personalized supplement (e.g., a supplement containing lactic acid bacteria, butyric acid bacteria, etc.) that improves the intestinal environment may be provided to the owner. In imaging tests, the coat, etc. is examined using images of the animal, and depending on the test results, a personalized supplement (a supplement containing vitamins, minerals, etc.) that improves the coat, etc. may be provided to the owner. Furthermore, when the owner terminal 210 receives a response to the medical interview (i.e., when a subjective test is conducted), depending on the test results, a personalized supplement containing herbal medicines, etc. may be provided to the owner.

[0104] FIG. 17 is a flowchart illustrating the processing steps executed by the server device 100 when providing a personalized supplement delivery service. The control unit 101 of the server device 100 accepts account registration from a pet owner through an online site established on the communication network NW (step S601). The control unit 101 issues an account ID to the pet owner who registers the account and accepts information about the pet owner and the pet owner's pets. The control unit 101 associates the accepted information with the account ID and stores it in the storage unit 102, thereby registering the account. The pet owner information includes the pet owner's name, address, telephone number, email address, and information about the primary and secondary medical institutions used by the pet owner. The pet owner's pet information includes the animal's type, sex, age, weight, and image. The animal information may also include the results of a consultation at the primary or secondary medical institution. The control unit 101 notifies the secondary medical institution of the pet owner and pet information accepted during account registration (step S602).

[0105] The secondary medical institution determines whether or not testing is necessary based on the animal information notified by the server device 100. If it determines that testing is necessary, the secondary medical institution notifies the owner of information on the necessary test items. The owner refers to the information notified by the secondary medical institution and, if necessary, requests testing by attaching a sample taken from the animal.

[0106] Notification from the secondary medical institution to the owner, and a request for testing from the owner to the secondary medical institution are made via the platform 10. That is, the control unit 101 of the server device 100 receives information on test items notified to the owner from the secondary medical institution and transmits the received information on test items to the owner terminal 210 (step S603). The control unit 101 of the server device 100 also receives a test request from the owner and transmits the received test request to the secondary medical institution (step S604). Note that samples required for testing are sent by mail or other means from the owner or the primary medical institution to the secondary medical institution. If images of the animal or answers to the medical interview required for the test are stored in the platform 10, these data are transmitted from the server device 100 to the secondary medical institution.

[0107] Secondary medical institutions accept test requests, receive the specimens necessary for the test, and then carry out the test. For example, if a request is received for a blood test using animal blood as a specimen, the secondary medical institution will carry out a blood test (clinical test) using the blood to examine the animal's condition. The same applies when requests are received for urine tests, genetic tests, imaging tests, etc., and the secondary medical institution will carry out the tests according to each test item.

[0108] After the test is completed, the secondary medical institution transmits the test results to the server device 10 so that the test results can be shared between the owner and the business operator. The test results are shared via the platform 10. Specifically, the control unit 101 of the server device 100 associates the test results received from the secondary medical institution with the owner's account ID and stores them in the memory unit 102 (step S605), and also transmits the test results to the owner terminal 210 used by the owner (step S606).

[0109] Secondary medical institutions will work with businesses to develop personalized supplements suited to individual animals. Secondary medical institutions can develop personalized supplements based on the animal information entered when registering an account (such as the animal's species, sex, age, weight, and images). In addition, if test results for the animal are available, secondary medical institutions can take the test results into consideration and develop personalized supplements to improve the animal's condition.

[0110] Information about the personalized supplements developed by the secondary medical institution is shared with the pet owner and the business operator via the platform 10. Specifically, the control unit 101 of the server device 100 receives the information about the personalized supplements sent from the secondary medical institution via the communication unit 103, and stores the received information about the personalized supplements in the storage unit 102 in association with the pet owner's account ID (step S607). The control unit 101 also notifies the pet owner terminal 210 of the information about the personalized supplements received from the secondary medical institution as information about recommended supplements (step S608).

[0111] When a pet owner wishes to purchase a personalized supplement jointly developed between a secondary medical institution and a business operator on a regular basis, the pet owner enters into a subscription contract with the business operator. To do so, the pet owner uses the pet owner terminal 210 to access the online site provided by the server device 100, logs in with their account ID, and then displays the subscription contract screen on the display unit of the pet owner terminal 210. The subscription contract screen accepts contract information, including the name and quantity of the personalized supplement the pet owner wishes to purchase on a regular basis, the name of the contract holder, the contract period, payment method, shipping address, shipping frequency, etc. The pet owner terminal 210 then transmits the contract information accepted through the contract screen to the server device 100.

[0112] The control unit 101 of the server device 100 receives the contract information transmitted from the owner terminal 210 via the communication unit 103 (step S609). A subscription contract is concluded when the provider accepts the contract information transmitted from the owner. The control unit 101 stores the contract information of the concluded subscription contract in the storage unit 102 in association with the owner's account ID (step S610).

[0113] The control unit 101 of the server device 100 references the contract information stored in the memory unit 102 and outputs a prescription request to the supplement manufacturer to prepare an individual supplement jointly developed between the secondary medical institution and the business (step S611). The individual supplement is manufactured by the supplement manufacturer in accordance with the prescription request and shipped to the owner at the shipping frequency (e.g., once a month, once a week, etc.) determined in the contract information.

[0114] As described above, in the second embodiment, by utilizing the platform 10, personalized supplements developed at secondary medical institutions such as university hospitals based on the test results of objective indicators such as blood tests, urine tests, and genetic tests can be provided to pet owners who have signed a subscription contract.

[0115] The embodiments disclosed herein should be considered in all respects as illustrative and not restrictive. The scope of the present invention is defined by the claims, not by the above meaning, and is intended to include all modifications within the meaning and scope of the claims.

[0116] Furthermore, the matters described in each embodiment can be combined with each other. Independent claims and dependent claims described in the claims can be combined with each other in any combination, regardless of the reference format. Furthermore, although the claims use a format in which a claim references two or more other claims (multiple claim format), this is not limited to this format. A multiple claim (multi-multi claim) that references at least one other multiple claim may also be used. [Explanation of symbols]

[0117] 1. Service provision system 10 Platform 100 Server device 101 Control section 102 Storage section 103 Communications Department 104 Operation section 105 Display section 210 Owner's Terminal 220 Veterinarian Terminal 230 Specialist Terminal 300 LLM Servers PG Service Delivery Program DB1 Management Database DB2 Supplement Database LM1 Health status estimation model LM2 Findings Estimation Model

Claims

1. A method for providing services to multiple users, including animal owners, veterinarians at primary care facilities, and specialists at secondary care facilities, via a platform for linking multiple types of veterinary medical services, comprising: collecting a plurality of case data including observation data of the animal, diagnostic data of a veterinarian who diagnosed the animal, and / or data of findings of the specialist on the observation data; Based on the collected case data, diagnostic support information is generated for veterinarians, transmitting the generated diagnostic support information to a terminal device used by the veterinarian; Information on supplements recommended for the animal diagnosed by the veterinarian is transmitted to a terminal device used by the owner. A service providing method in which processing is executed by a computer system.

2. generating a first learning model using as training data a dataset including the collected observation data and the veterinarian's diagnosis data, the first learning model being configured to predict the health status of an animal when the observation data is input; When observation data is received from the terminal device used by the owner, the received observation data is input into the first learning model and a calculation is performed using the first learning model; The estimated result regarding the health condition of the animal obtained from the first learning model is transmitted to a terminal device used by the owner. The service providing method according to claim 1 , wherein the processing is executed by the computer system.

3. generating a first learning model using as training data a dataset including the collected observation data and the veterinarian's diagnosis data, the first learning model being configured to predict the health status of an animal when the observation data is input; When observation data is received from the terminal device used by the owner, the received observation data is input into the first learning model and a calculation is performed using the first learning model; transmitting the estimation result regarding the animal's health condition obtained from the first learning model to a terminal device used by a veterinarian at the primary medical institution; A diagnosis result of a doctor who refers to the estimation result is transmitted to a terminal device used by the owner. The service providing method according to claim 1 , wherein the processing is executed by the computer system.

4. The observation data includes at least one of image data, vital data, sound data related to the animal's cry, olfactory data, and data on responses to a medical interview. The service providing method according to claim 1 .

5. receiving speech data including the content of the consultation from the owner; generating response data corresponding to the received utterance data by referring to the case data; The generated response data is returned to the terminal device used by the owner. The service providing method according to claim 1 , wherein the processing is executed by the computer system.

6. extracting a query from the owner from the speech data; generating a response sentence to the extracted query using a language model that references the case data and literature data related to animal diagnosis; Response data including the generated response sentence is returned to the terminal device used by the owner. The service providing method according to claim 5 , wherein the processing is executed by the computer system.

7. receiving speech data including the consultation content from the veterinarian; generating response data corresponding to the received utterance data by referring to the case data; The generated response data is returned as diagnostic support information to the terminal device used by the veterinarian. The service providing method according to claim 1 , wherein the processing is executed by the computer system.

8. extracting the veterinarian's query from the speech data; generating a response sentence to the extracted query using a language model that references the case data and literature data related to animal diagnosis; The answer data including the generated answer sentence is returned to the terminal device used by the veterinarian. The service providing method according to claim 7 , wherein the processing is executed by the computer system.

9. generating a second learning model configured to estimate the specialist's findings on the observed image when the observation data is input, using a dataset including the collected observation data and the specialist's findings on the observation data as training data; When observation data is received from the terminal device used by the veterinarian, the received observation data is input into the second learning model and a calculation is performed using the second learning model; generating diagnostic support information for the veterinarian based on the estimation result of the second learning model; The service providing method according to claim 1 , wherein the processing is executed by the computer system.

10. the observation data includes image data relating to a medical image; Based on the estimation result of the second learning model, an interpretation report for the medical image is generated as the diagnostic support information. The service providing method according to claim 9.

11. the observation data includes at least one of image data, vital data, sound data relating to the animal's calls, olfactory data, and data on responses to a medical interview; Based on the estimation result of the second learning model, information regarding at least one of a treatment method, a therapeutic drug, and a supplement for the animal is generated as the diagnostic support information. The service providing method according to claim 9.

12. Displaying a purchase contract screen for concluding a purchase contract for the supplement on a terminal device used by the pet owner, Outputting a shipping instruction for the supplement in accordance with the purchase contract concluded through the purchase contract screen The service providing method according to claim 1 , wherein the processing is executed by the computer system.

13. The case data includes information on supplements recommended by the veterinarian or the specialist depending on the case of the animal, Displaying information about supplements recommended by the veterinarian or the specialist on a terminal device used by the owner The service providing method according to claim 12, wherein the processing is executed by the computer system.

14. The supplement is a personalized supplement developed for the animal at the secondary care facility based on samples obtained from the animal. The service providing method according to claim 12.

15. A service providing device that provides services to multiple users, including animal owners, veterinarians at primary medical institutions, and specialists at secondary medical institutions, via a platform for linking multiple types of veterinary medical services, one or more processors; the one or more processors: collecting a plurality of case data including observation data of the animal, diagnostic data of a veterinarian who diagnosed the animal, and / or data of findings of the specialist on the observation data; Based on the collected case data, diagnostic support information is generated for veterinarians, transmitting the generated diagnostic support information to a terminal device used by the veterinarian; Information on supplements recommended for the animal diagnosed by the veterinarian is transmitted to a terminal device used by the owner. Service providing device.

16. A computer program for causing a computer to execute a process for providing services to multiple users, including animal owners, veterinarians at primary care institutions, and specialists at secondary care institutions, via a platform for linking multiple types of veterinary medical services, the computer program comprising: collecting a plurality of case data including observation data of the animal, diagnostic data of a veterinarian who diagnosed the animal, and / or data of findings of the specialist on the observation data; Based on the collected case data, diagnostic support information is generated for veterinarians, transmitting the generated diagnostic support information to a terminal device used by the veterinarian; Information on supplements recommended for the animal diagnosed by the veterinarian is transmitted to a terminal device used by the owner. A computer program for causing the computer to execute a process.

Images