WO2020213843A1

WO2020213843A1 - User-customized medical information provision system and operating method therefor

Info

Publication number: WO2020213843A1
Application number: PCT/KR2020/003867
Authority: WO
Inventors: 이정의
Original assignee: 이정의
Priority date: 2019-04-19
Filing date: 2020-03-20
Publication date: 2020-10-22
Also published as: KR102088980B1

Abstract

The present invention relates to a user-customized medical information provision system using artificial intelligence technology, and an operating method therefor, the medical information provision system comprising: an information reception unit for receiving user information and symptom information from a user terminal; a diagnosis unit, which matches, with information stored in big data, a text extracted from the symptom information received in the information reception unit, so as to identify a diagnosis name according to the symptom matched with at least a preset reference matching rate; and an information provision unit for providing, to the user terminal, the diagnosis name identified by the diagnosis unit, and the information which is required for a treatment and includes at least one recommended hospital or medical office, and thus the present invention resolves the problem in which overtreatment by medical staff and indiscriminate medical shopping by a medical service user increase, thereby enabling the health of a medical consumer to be actively managed.

Description

User-customized medical information provision system and its driving method

The present invention relates to a user-customized medical information providing system and a driving method thereof, and more particularly, to a user-customized medical information providing system and a driving method thereof to which deep learning technology of artificial intelligence is applied.

Artificial intelligence refers to intelligence created by machines or software (SW) capable of cognitive abilities, reasoning abilities, learning abilities, and comprehension abilities, which are considered uniquely possessed by humans. It is collectively referred to as the field of science that studies how to make computers or computer SW that can perform intelligent functions. It mainly includes pattern recognition, expert system, natural language processing, machine learning and automatic control technology.

A platform that provides services related to securities investment and transactions, and artificial intelligence technology that uses artificial intelligence to grasp various information and provide customized investment-related information to service users have been introduced.

In addition, artificial intelligence that analyzes a vast amount of real-time data in asset management and advises the direction of asset management, or technology for attorney's work automation that uses artificial intelligence to create a payment order report has been proposed.

In addition, as a system developed through the combination of cloud computing, big data, artificial intelligence technology, and traditional medical industry, a technology capable of finding problems based on symptoms through collection and analysis of large amounts of medical data and specialized literature has been proposed. It is possible to exchange with the patient on behalf of the doctor based on the voice recognition function, collect the extension information from the patient, make a diagnosis based on the information previously collected-diagnosed by the doctor, and submit an opinion. Accordingly, there is an effect of saving the doctor's time and improving the efficiency of the examination process.

On the other hand, with the development of large-capacity data transmission and processing technology, the number of contents that each individual can access is increasing exponentially. In particular, with the introduction of high-speed communication networks and the mass increase of various multimedia devices such as smartphones, tablets, netbooks, IP, and TVs, users can enjoy numerous contents regardless of time and place.

However, as the amount of content increases, it takes more time and effort for users to find the information they want. Therefore, a method of selecting and recommending content that satisfies the user has emerged.

Such conventional recommendation methods include a simple recommendation method using content sales volume and recommendation information of buyers, and a customized recommendation method using preference genre information or preference category information input by each user, or purchase history of each user.

However, in the case of a simple recommendation method based on sales volume, etc., there is a problem that recommendation accuracy is degraded because it cannot reflect the user's personal taste.

(Patent Document 1) KR10-0739570 B1

(Patent Document 2) KR10-1525576 B1

(Patent Document 3) KR10-0538573 B1

The present invention has been derived from such a technical background, and an object thereof is to solve a problem in which the imbalance of medical information is not resolved compared to the speed of medical technology development.

In addition, it intends to provide a user-customized medical information provision system and its driving method that can contribute to the promotion of public health and prevent excessive medical treatment and medical shopping.

The present invention for achieving the above object includes the following configuration.

That is, the medical information providing system according to an embodiment of the present invention matches the text extracted from the information receiving unit receiving user information and symptom information from the user terminal and symptom information received by the information receiving unit with information stored in big data. Includes a diagnostic unit that identifies a diagnosis name according to a symptom matched by a set reference matching rate or higher, and an information providing unit that provides information necessary for treatment including at least one recommended hospital or department of treatment to the user terminal, and the diagnosis name identified by the diagnosis unit. Characterized in that.

On the other hand, the driving method of the user-customized medical information providing system is that the user terminal, which is mounted and driven by a medical information providing service app, is provided by the user through the medical information providing service app. Receiving user information and symptom information including at least one of a biographical history and working environment, a medical information providing system receiving the user information and symptom information from the user terminal, and a text extracted from the received symptom information. Matching the information stored in the data to identify a diagnosis name according to a symptom that matches a preset reference matching rate or higher, and the medical information providing system provides the identified diagnosis name to a user terminal, and at least one hospital or medical department necessary for treatment It characterized in that it comprises a step of recommending.

According to the present invention, medical consumers can take the lead in health care by solving the problem of increasing medical care by medical staff and indiscriminate medical shopping by medical service users by accurately delivering medical knowledge about what medical treatment and treatment are necessary. The effect that enables it is derived.

In addition, according to the medical information providing service system and its driving method according to an embodiment of the present invention, by providing customized medical information that medical consumers (patients) can easily access, improving the quality of life and at the same time social It is expected to bring about reduction of indirect capital.

In addition, by presenting treatment methods for similar cases based on health care big data, a user-customized medical information provision system and its driving method that can reduce medical costs by preventing excessive treatment that may occur due to unbalanced medical information are provided. There is an effect it can provide.

Furthermore, according to the present invention, not only can it be possible to provide precise medical services by providing user-specific information based on big data analysis based on information such as the user's medical history, medical history, family history, etc., but also the user's specific information (existing medical history, It is possible to prevent medical accidents due to lack of user information by collecting information such as biographical history, side effects, etc., and transmitting the user's information when the user consults with a health care provider.

1 is a block diagram showing the configuration of a user-customized medical information providing system according to an embodiment of the present invention;

2 is a block diagram showing in more detail the configuration of a medical information providing service server according to an embodiment of the present invention;

3 is a flowchart illustrating a method of driving a system for providing customized medical information according to an embodiment of the present invention;

4 is an exemplary diagram for explaining an embodiment of a driving method of a system for providing customized medical information according to an embodiment of the present invention.

It should be noted that the technical terms used in the present invention are only used to describe specific embodiments, and are not intended to limit the present invention. In addition, the technical terms used in the present invention should be interpreted as generally understood by those of ordinary skill in the technical field to which the present invention belongs, unless otherwise defined in the present invention, and is excessively comprehensive. It should not be construed as a human meaning or an excessively reduced meaning.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram showing the configuration of a user-customized medical information providing system according to an embodiment of the present invention.

As can be seen from FIG. 1, a user-customized medical information providing system according to an embodiment of the present invention includes a user terminal 10, a medical information providing service server 20, a medical information server 30, and a location identification server 40. Includes.

The user terminal 10 is an IP-allocated terminal and can perform network communication through the Internet or the like. For example, as a wireless communication device with guaranteed portability and mobility, navigation, PCS (Personal Communication System), GSM (Global System for Mobile communications), PDC (Personal Digital Cellular), PHS (Personal Handyphone System), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), Wibro (Wireless Broadband Internet) terminal, smartphone, smart phone It may include all kinds of handheld-based wireless communication devices such as a smartpad and a tablet PC.

In addition, a desktop PC, a slate PC, a notebook computer, and a portable multimedia player (PMP) may be applicable. Of course, the terminal to which the present invention is applicable is not limited to the above-described types, and it is natural that all terminals capable of communicating with external devices may be included.

In one embodiment, the user terminal 10 is driven by being equipped with a medical information providing service-only app, and through the medical information providing service app, the user's age, gender, pregnancy status, specifics, family history, medical history, biographies, work User information and symptom information including at least one of the environments are input.

In addition, a variety of information that can identify the user's overall physical condition is received, such as drug allergic reactions, living environment, dietary habits, and drug information.

At this time, an app dedicated to providing medical information may provide a convenient user interface for inputting symptoms. For example, a screen for selecting a part of the body part is provided, and a screen for selecting in more detail the symptoms that may occur in the body part selected step by step is further provided.

The user terminal 10 includes a communication module that performs communication with the medical information providing service server 20 including a communication module. At this time, the communication module is a communication module supporting a short-range wireless communication method between devices as well as a communication module supporting a communication method using a communication network (for example, a mobile communication network, wired Internet, wireless Internet, and broadcasting network) that the network may include. Can also be included.

The medical information providing service server 20 includes a computer device or a server device that performs primary diagnosis. In one embodiment, since the medical information providing service server 20 needs to process a vast amount of data, it may be implemented as a parallel computing system or a cloud computing environment in which at least one computer is connected in parallel.

In addition, the medical information providing service server 20 uses various algorithms of artificial intelligence. The medical information providing service server 20 is connected to the user terminal 10 through a network through a wired/wireless communication network to transmit and receive data.

Here, the network is one of networks such as PAN (personal area network), LAN (local area network), CAN (campus area network), MAN (metropolitan area network), WAN (wide area network), BBN (broadband network), and the Internet. It may include one or more of any network.

In addition, the network may include any one or more of a network topology including a bus network, a star network, a ring network, a mesh network, a star-bus network, a tree, or a hierarchical network, but is not limited thereto. Does not.

The medical information providing service server 20 receives user information and symptom information from the user terminal 10, matches the text extracted from the received symptom information with information stored in big data, and matches a symptom that matches a preset reference matching rate or higher. The diagnosis name according to is identified, the identified diagnosis name is provided to the user terminal, and at least one hospital or medical department required for treatment is recommended.

Here, the medical information providing service server 20 learns in a supervised learning method. Supervised learning is a method of machine learning to infer a function from training data. A label for data-a method of learning in the form of data and labels given an explicit correct answer. to be. After training with the training dataset, it is possible to measure how accurately the learned algorithm predicts using the unlabeled test dataset.

In this case, the learning dataset used by the medical information providing service server 20 may include, for example, personal disease history and personal medical history information among the user's yearly medical data provided by the KNHIS database.

In the pre-processing process, values of personal disease history and personal medical history are set as a learning dataset based on the user's medical data for each year, and learned through supervised learning. When a user inputs medical data through an app dedicated to the medical information providing service of the user terminal, the values of the personal disease history and personal medical history are set as input values and relearned.

In one embodiment, the medical information providing service server 20 employs an artificial intelligence system. Artificial intelligence systems include control structures, knowledge bases, databases, and inference engines.

The control structure translates the rules of various algorithms with a rule translator. The knowledge base contains various artificial intelligence algorithms as rules. In addition, the inference engine may classify data in the database and classify input images using rules translated by an expert rule translator.

In the artificial intelligence system of the medical information providing service server 20, the knowledge base includes rules for detecting medical lesions, and the database includes a second diagnosis result in which the expert diagnosis result is added to the artificial intelligence-based first diagnosis result. Re-diagnosis and re-learning are performed using together.

In addition, the medical information providing service server 20 receives the final diagnosis and classification result from the user terminal 10 after providing the second diagnosis result to the user terminal 10 and stores it in the database 25 as a final decision. . In this case, the final diagnosis and classification result may be a prescription image received after a user directly visits a hospital. Alternatively, prescription information issued to the user may be directly received from the hospital. In this case, preferably, a process of obtaining consent from the user or the hospital for provision of prescription information should be preceded.

The final diagnosis and classification results can be stored and stored together with the AI analysis results, and then used for re-learning and performance evaluation. In this case, relearning includes a process of labeling and using the input value as the result of the final diagnosis and classification that has been finally confirmed by the user.

In one embodiment, the medical information server 30 stores symptom information on various diseases including big data. The medical information providing service server 20 may be interlocked with the medical information server 30.

The medical information server 30 includes health care big data. For example, it may include information obtained by extracting the search results of the portal site for recommendation treatment by symptoms and converting them into a database. In addition, it may include information obtained by converting treatment by department, search results for examination items, or treatment history of doctors into a database.

In one embodiment, the medical information server 30 includes a KNHIS database.

In another embodiment, the medical information server 30 may collect health care big data from a portal site or a medical institution server such as a hospital or pharmacy using web crawling.

The location determination server 40 is interpreted to encompass a technical configuration for identifying the location information of the user terminal 10. In an exemplary embodiment, the location detection server 40 may receive a GPS signal from the user terminal 10 to obtain location information of the user terminal 10. Alternatively, the location determination server 40 may be implemented as a communication company server. That is, the location determination server 40 is interpreted to encompass a technical configuration capable of identifying the current location information of the user terminal 10 in various technical methods.

In one embodiment, when a request for location information of the user terminal 10 is received from the medical information providing service server 20, the location identification server 40 identifies the current location information of the user terminal 10 and provides medical information. It is provided to the service server 20.

The medical information providing service server 20 according to an aspect identifies the location information of the user terminal 10 from the location detection server 40, and uses the location information of the identified user terminal 10 to provide medical treatment. Recommend information.

According to a characteristic aspect of the present invention, the medical information providing service server 20 provides a diagnosis name according to symptom information received from the user terminal 10 and a prescription input from the recommended hospital or treatment department based on deep learning technology. By comparing and calculating the accuracy of the diagnosis name (disease code, treatment code), the diagnosis name according to the symptom is learned according to symptom information received from the user terminal 10.

In other words, after the user actually visits a hospital or treatment department, it can evolve more in grasping the diagnosis name by learning a prescription issued by a doctor.

Deep learning is a set of machine learning algorithms that attempt a high level of abstraction through a combination of several nonlinear transformation methods. In this case, abstraction refers to the work of summarizing key contents or functions in a large amount of data or complex data.

This is a field of machine learning that teaches human thinking to computers. It can dramatically accelerate the speed of neural network computation with the advent of probability vectors, algorithm parallelism, and GPUs that predict results based on weights based on sufficient data.

In addition, according to an additional aspect of the present invention, the medical information providing service server 20 provides the same symptom and diagnosis name as the previous user based on the diagnosis name according to the symptom identified by the symptom information received from the user terminal 10. Provide more information on the prescription history of other patients.

At this time, the user's age or whether or not other diseases have occurred may be investigated together. For example, in the case of a runny nose from throat, information on changes in medications administered for the disease is provided based on treatment history information stored in big data.

At this time, the treatment history information stored in the big data is prescription information directly registered by another person in the medical information providing service server 20 according to an embodiment of the present invention, or by web crawling that is personally shared through web pages such as cafes and blogs. It may be collected information. Also, it may be based on information obtained by converting a search result of a portal site and recommendation treatment for each symptom collected by the medical information server 30 into a database.

Accordingly, it is possible for the medical staff to make a more thoughtful diagnosis to each patient, and the patient particle can go through a verification procedure based on the actual treatment history information of other patients who complained of the same symptoms before and can present their opinions. It becomes possible to receive medical treatment.

In one embodiment, the medical information providing service-only app of the user terminal 10 includes hospital information recommended to the user or the name information on the medical department or the identified disease according to the information received from the medical information providing service server 20. It can be printed as a list.

In addition, a reservation request may be received by selecting information of a hospital, a medical department, or a person in charge from the user. When the user terminal 10 transmits the information of the hospital or department in charge selected by the user to the medical information providing service server 20, the medical information providing service server 20 is a doctor selected by the user through the selected hospital homepage or treatment department homepage. It is possible to request a medical appointment for and further upload the user information identified based on the information received from the user terminal 10 to the hospital homepage or the treatment department homepage.

At this time, at the same time as the user's appointment, the user information including symptom information already entered by the user or at least one of age, gender, pregnancy status, peculiarities, family history, medical history, medical history, and work environment is delivered to the hospital or department of the appointment. I can. Accordingly, it is possible to perform a simple interview, thereby simplifying the cumbersome process of having to go through each interview in the first hospital.

In addition, it is possible to receive user information and set it as a variable value, and by using this, it is possible to provide more personalized medical information by deriving the difference between the previously learned personal disease history and the treatment department by symptom according to the personal medical history. And you can see which variable values had more influence.

2 is a block diagram showing in more detail the configuration of a medical information providing service server according to an embodiment of the present invention.

The medical information providing service server according to an embodiment includes a communication unit 200, an information receiving unit 210, a diagnosis unit 220, an information providing unit 230, and a learning unit 240.

The communication unit 200 performs data communication with the user terminal 10, the medical information server 30, and the location server 40. The communication unit 200 includes not only a communication module supporting a communication method using a communication network (for example, a mobile communication network, wired Internet, wireless Internet, broadcasting network), but also a short-range wireless communication method between devices. Modules can also be included.

The information receiving unit 210 receives user information and symptom information from the user terminal 10 through the communication unit 200. The information receiving unit 210 is provided with an app for providing medical information service through the communication unit 200 from the user terminal 10, which is operated by the user through an app dedicated to providing medical information. , User information and symptom information including at least one of medical history, biographical history, and work environment are received. In addition, a variety of information that can identify the user's overall physical condition is received, such as drug allergic reactions, living environment, dietary habits, and drug information.

The diagnosis unit 220 identifies a diagnosis name according to a symptom matching a predetermined reference matching rate or higher by matching text extracted from symptom information received by the information receiving unit 210 with information stored in big data.

The information providing unit 230 provides the user terminal 10 with information necessary for treatment including a diagnosis name identified by the diagnosis unit 220 and at least one recommended hospital or department.

The learning unit 240 compares the diagnosis name according to symptom information received from the user terminal 10 and a prescription input from the recommended hospital or treatment department based on deep learning technology, and the accuracy of the diagnosis name (disease code, treatment code) A diagnosis name according to symptoms is learned according to symptom information received from the user terminal 10 by calculating.

In one aspect, the learning unit 240 may learn through supervised learning by setting personal disease history and personal medical history information among the user's yearly medical data stored in big data as a learning dataset.

3 is a flowchart illustrating a method of driving a system for providing customized medical information according to an embodiment of the present invention.

First, a user terminal that is equipped with a medical information providing service app installed and operated is a user including at least one of age, gender, pregnancy, special information, family history, medical history, biographical history, and work environment from the user through the medical information providing service app. Information and symptom information are input (S300).

In addition, the medical information providing service server receives user information and symptom information from the user terminal (S310), and matches the text extracted from the received symptom information with information stored in big data, according to symptoms matching a preset reference matching rate or higher. The diagnosis name is identified (S320).

According to an aspect of the present invention, the medical information providing service server determines the location information of the user terminal (S330).

Thereafter, the medical information providing service server provides the identified diagnosis name to the user terminal, and recommends at least one hospital or department that is necessary for treatment and capable of treatment by using the location information of the user terminal (S340).

In addition, in one embodiment, the medical information providing service server compares the diagnosis name primarily identified based on the deep learning technology with the prescription input from the recommended hospital or department, and calculates the accuracy of the diagnosis name received from the user terminal. The diagnosis name according to the symptoms is learned according to the symptom information (S350).

According to an additional aspect of the present invention, the medical information providing service server further provides information on the prescription history of another patient for the same symptom and diagnosis name previously based on the diagnosis name according to the user's symptom input from the user (S360).

In this case, the user's age or whether or not other diseases are onset may be investigated together to provide a prescription history of another patient corresponding to a similar situation. For example, in the case of a runny nose from throat, information on changes in medications administered for the disease is provided based on treatment history information stored in big data.

Accordingly, it is possible for the medical staff to make a more thoughtful diagnosis to each patient, and the patient particle can go through a confirmation procedure based on the treatment history information of other patients before and can provide an opinion, making it possible to receive highly reliable treatment. .

In addition, in an embodiment, when a reservation request is received from an app dedicated to the medical information providing service of the user terminal (S370), the medical information providing service server performs a medical reservation with at least one of the recommended hospitals or treatment departments (S380). In addition, it is possible to recommend more specialists in the field. Then, according to the user's reservation request, a medical appointment can be made to the recommended specialist.

4 is an exemplary view for explaining an embodiment of a driving method of a system for providing user-customized medical information according to an embodiment of the present invention.

First, the user may sign up and log in as a member through an app dedicated to providing medical information service running on the user terminal 10 (S400). When registering as a member, unique information including at least one of the user's age, gender, pregnancy status, special details, family history, medical history, biographical history, and work environment may be input in advance (S420, S422).

And when a user logs in, the unique information entered at the time of membership registration can be automatically recognized. However, it is not limited thereto.

The medical information providing service server 20 requests health care big data from the medical information server 30 (S410), and collects health care information according to the request result (S415). Then, it is stored in a database provided in the medical information providing service server 20.

At this time, the user enters his or her basic symptom information after logging in through a dedicated app for providing medical information. Upon logging in, the medical information providing service server 20 may grasp the unique information input at the time of membership registration, and further grasp the additionally input symptom information.

In addition, it is also possible to receive a variety of information that can identify the user's overall physical condition, such as drug allergic reactions, living environment, eating habits, and drug information being taken.

The medical information providing service server 20 extracts metadata from the health care big data obtained from the medical information server 30 and the basic symptom information text of the user received from the user terminal 10 (S424).

In addition, the medical information providing service server 20 matches the text extracted from the received symptom information with the information stored in the health care big data, and extracts data matching a preset reference matching rate or higher.

The medical information providing service server 20 extracts data whose matching rate with text included in the user information is greater than or equal to the reference matching rate among health care big data and transmits it to the user terminal 10 (S426).

Then, the user terminal 10 displays information received from the medical information providing service server 20 on a screen, and receives selection of additional information from the user through an app dedicated to the medical information providing service (S430). Here, the additional information may be information for identifying more detailed symptoms of the user.

Then, the user terminal 10 transmits the selection information received from the user through the medical information providing service dedicated app to the medical information providing service server 20 (S432).

The medical information providing service server 20 stores and learns selection information received from the user terminal 10 (S434), and then extracts user preference information based on the learned information (S436).

In this case, the preference information may be part of information classified by criteria such as a location or characteristic of a medical institution that the user prefers.

The medical information providing service server 20 transmits the extracted user's preference information to the user terminal 10, and the user inputs user symptoms in detail through an app screen dedicated to the medical information providing service of the user terminal 10 ( S440).

In addition, the user may transmit a customized treatment and recommendation request to the medical information providing service server 20 through the medical information providing service dedicated app screen of the user terminal 10 (S450).

The medical information providing service server 20 extracts treatment department information for each user symptom based on the health care big data collected from the medical information server 30 (S452). In addition, the medical information providing service server 20 may further extract treatment content information for each treatment department based on the health care big data collected from the medical information server 30 (S454).

Thereafter, the medical information providing service server 20 extracts user-customized treatment department information based on the user information based on the extracted information (S456).

Additionally, the medical information providing service server 20 requests the location-based information of the user terminal from the location determination server 40 (S460).

Then, the location determination server 40 transmits the user location-based information to the medical information providing service server 20 (S462).

In addition, the medical information providing service server 20 generates a user location-based extraction information list by using the identified user location information (S464). In other words, it is possible to include in the list hospitals in a short distance that can be reached within the treatment hours. Alternatively, it is possible to extract hospitals and departments within a radius of several kilometers (km) and include them in the list.

In addition, the medical information providing service server 20 provides the user terminal 10 with a list including information on a recommended hospital or department generated by extracting user-customized treatment department information.

That is, the medical information providing service server 20 provides the identified diagnosis name to the user terminal 10, and recommends at least one hospital or department that is necessary for treatment and capable of treatment by using the location information of the identified user terminal. will be.

After that, after visiting the hospital, the user registers a prescription issued from the actual hospital through the user terminal 10 (S470). In this case, the prescription may be provided as a photographed image from the user terminal 10.

Then, the user terminal 10 requests a prescription registration to the medical information providing service server 20 (S472). At this time, when a prescription registration request is received from the user terminal 10, the medical information providing service server 20 may request and receive prescription information for the user from a hospital or pharmacy server. Alternatively, the medical information providing service server 20 may receive a prescription image file taken directly from the user terminal 10 at the same time as the prescription registration request.

The medical information providing service server 20 stores the obtained prescription in the database (S474).

By storing the prescription in a database, it is possible to provide the prescription information registered by the user through the fetch function. At this time, it is also possible to provide the prescription confirmation by classification by symptom or by department. Accordingly, the user can grasp his or her prescription history in the past.

In addition, the medical information providing service server 20 may provide prescription information prescribed by another person having the same disease or symptom as the user, even if the prescription is not necessarily issued by the user.

Thereafter, the medical information providing service server 20 according to a characteristic aspect of the present invention compares the diagnosis name primarily identified based on the deep learning technology with a prescription input from the recommended hospital or medical department to determine the accuracy of the diagnosis name. By calculating, a diagnosis name according to a symptom is learned according to symptom information received from the user terminal 10 (S480).

More personalized medical information can be provided by setting the user's personal information including at least one of age, gender, pregnancy status, special matters, family history, medical history, biographical history, and work environment as a variable value and re-learning.

In providing customized information to the user terminal 10 based on the learned content afterwards (S490), it is possible to identify a diagnosis of a symptom with higher accuracy and recommend a hospital and a department of treatment.

The above-described method may be implemented as an application or in the form of program instructions that may be executed through various computer components and recorded in a computer-readable recording medium. The computer-readable recording medium may include program instructions, data files, data structures, etc. alone or in combination.

The program instructions recorded in the computer-readable recording medium may be specially designed and constructed for the present invention, and may be known and usable to those skilled in the computer software field.

Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical recording media such as CD-ROMs and DVDs, and magnetic-optical media such as floptical disks. media), and a hardware device specially configured to store and execute program instructions such as ROM, RAM, flash memory, and the like.

Examples of the program instructions include not only machine language codes such as those produced by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like. The hardware device may be configured to operate as one or more software modules to perform processing according to the present invention, and vice versa.

Although the above has been described with reference to embodiments, those skilled in the art will understand that various modifications and changes can be made to the present invention without departing from the spirit and scope of the present invention described in the following claims. I will be able to.

Claims

An information receiver configured to receive user information and symptom information through an app dedicated to providing medical information service of the user terminal;

A diagnosis unit that matches the text extracted from symptom information received by the information receiving unit with information stored in big data to determine a diagnosis name according to a symptom matching a predetermined reference matching rate or higher; And

Including; an information providing unit for providing information necessary for treatment, including the diagnosis name identified by the diagnosis unit and at least one recommended hospital or department, to a user terminal; and

Based on the deep learning technology, the diagnosis name determined by the diagnosis unit is compared with the prescription issued by the recommended hospital or treatment department, and the accuracy of the diagnosis name is calculated to learn the diagnosis name according to the symptoms according to the symptom information received from the user terminal. ,

At least one of gender, age, residence, medical subject code, examination institution type code, past medical history code, family history, special details, medical history, biographical history, and work environment among the medical data of the National Health Insurance Corporation (KNHIS) big data by year Set the training dataset including the variable value of,

Including; a learning unit that learns the diagnosis name according to the symptom through supervised learning so that the learning result matches the symptom name of the big data of the National Health Insurance Service (KNHIS),

The diagnosis unit is a user-customized medical information providing system using artificial intelligence that identifies a disease code included in a diagnosis name based on information stored in health care big data.
The method of claim 1,

The diagnosis unit inputs at least one of the user's gender, age, residence, medical subject code, examination institution type code, past medical history code, family history, special details, medical history, biographical history, and work environment received from the user terminal to the information receiving unit. Receive and derive a diagnosis name,

The learning unit,

A user-customized medical information providing system using artificial intelligence that improves the accuracy of the diagnosis name with a test dataset that compares the diagnosis name derived from the diagnosis unit with a prescription issued by the hospital or medical department to check whether they match.
The user terminal, which is equipped with a medical information providing service app, is installed and operated from the user through the medical information providing service app, from the user to the user's gender, age, place of residence, medical treatment subject code, examination institution type code, past medical history code, family history, special information, Receiving user information and symptom information including at least one of a medical history, a medical history, and a work environment;

Receiving, by a medical information providing system, the user information and symptom information from the user terminal, and deriving a disease code included in an expected diagnosis name from the received information; And

Including, by the medical information providing system, providing a disease code included in the identified diagnosis name to a user terminal, and recommending at least one hospital or medical department in which the identified diagnosis name can be treated; and

The medical information providing system calculates the accuracy of the diagnosis name by comparing the identified diagnosis name with the prescription input from the recommended hospital or medical department based on deep learning technology, and according to the symptom information received from the user terminal. Further comprising; learning the diagnosis name according to,

The learning step,

At least one of gender, age, residence, medical subject code, examination institution type code, past medical history code, family history, special details, medical history, biographical history, and work environment among the medical data of the National Health Insurance Corporation (KNHIS) big data by year Set the training dataset including the variable value of,

The diagnosis name according to the symptoms is learned through supervised learning so that the learning results match the symptom names of the big data of the National Health Insurance Corporation (KNHIS).

The deriving step is a driving method of a user-customized medical information providing system using artificial intelligence that identifies a disease code included in a diagnosis name based on information stored in health care big data.
The method of claim 3,

The deriving step,

At least one of the user's gender, age, place of residence, treatment subject code, examination institution type code, past medical history code, family history, special details, medical history, biographical history, and work environment received from the user terminal to the information receiving unit of the medical information providing system It receives the input and derives the diagnosis name,

The learning step,

User-customized medical information providing system using artificial intelligence that improves the accuracy of the diagnosis name with a test dataset that compares the diagnosis name derived from the diagnosis unit of the medical information providing system with a prescription issued by the hospital or medical department to check whether they match Method of driving.