KR20210056598A - Method and system for providing medical data collection and analyzing service based on machine learning - Google Patents

Abstract

According to the present invention, a method of providing a medical data collection and analysis service based on machine learning includes: (a) collecting a plurality of medical data, and performing pre-processing and classification on the collected data; (b) constructing a medical learning model by performing machine learning with reference to the classified medical data; (c) providing a user interface for a medical data collection and analysis platform to a user terminal to allow a user to randomly input a plurality of medical data for a project to be analyzed as desired, and collecting the medical data input from the user terminal; (d) removing or correcting some data by performing an optimization operation on the medical data; and (e) providing an output value to the user terminal by inputting the collected medical data into the constructed medical learning model, wherein the medical data includes medical-related institution data including medical treatment data and clinical information data, and medical academic data including medical thesis data. Accordingly, an output value including a disease name, a treatment method, and disease prognosis prediction information, which is highly related to the project input by the user, is provided.

Description

Machine learning-based medical data collection and analysis service provision method and system {METHOD AND SYSTEM FOR PROVIDING MEDICAL DATA COLLECTION AND ANALYZING SERVICE BASED ON MACHINE LEARNING}

The present invention relates to a method and system for providing medical data collection and analysis services based on machine learning.

With the recent development of artificial intelligence technology, big data technology that extracts useful information by processing a lot of data is attracting attention.

For example, in the pharmaceutical industry, artificial intelligence is being applied for efficient drug development. In the medical field, artificial intelligence technology and big data technology are attracting attention as technologies for medical diagnosis and prediction.

The conventional medical information field is being divided into a medical field and a research field to develop. For example, patient care data that stores patient medical records for each medical information field, clinical information data that stores test results of subjects participating in clinical trials, medical paper data, patient health checkup data, hospitals My electronic medical chart (EMR, Electronic Medical Record) is managed as a database through a representative information management system in each field.

However, in the current medical field, the above-described various medical data are not classified and managed according to related categories, and for this reason, it takes a lot of time to search for medical documents and medical data within one correlated category. There is a problem of becoming.

In this regard, Korean Patent Publication No. 10-2012-0004611 discloses a medical test result data management system and a management method.

The present invention is to solve the problems of the prior art described above, and provides a user interface through which a user can input a plurality of medical data for a project to be analyzed, and an output value analyzed by a medical learning model is transmitted to a user terminal. We intend to provide a medical data collection and analysis service system based on machine learning provided by

However, the technical problem to be achieved by the present embodiment is not limited to the technical problem as described above, and other technical problems may exist.

As a technical means for achieving the above-described technical problem, the method for providing medical data collection and analysis service based on machine learning according to the first aspect of the present invention includes (a) collecting a plurality of medical data and pre-processing the collected data. And performing classification; (b) constructing a medical learning model by performing machine learning by referring to the classified medical data; (c) Provides a medical data collection and analysis platform user interface to the user terminal, allowing the user to randomly input multiple medical data for the project to be analyzed as desired, and to collect a plurality of medical data input from the user terminal. step; (d) removing or correcting some data by performing an optimization operation on a plurality of medical data; And (e) inputting a plurality of collected medical data to the established medical learning model and providing an output value to the user terminal, wherein the medical data includes medical-related institutional data and medicine including medical treatment data and clinical information data. Includes medical academic data including thesis data.

The step (a) includes the steps of performing de-identification processing of information corresponding to personal information among the collected data to be removed from the collection object or processed so that humans cannot recognize it; Labeling the types of collected data; And classifying the labeled data by category.

In step (b), machine learning is performed by extracting keywords from medical data of each category, setting keywords related to symptoms or patient conditions as input values, and setting keywords related to disease results, treatment methods, or clinical results as output values. However, machine learning is performed by determining a machine learning algorithm based on the category or type of medical data, and selecting an optimal algorithm through a fine tuning process that automatically performs a combination between a plurality of machine learning algorithms.

Step (c) includes receiving a selection result for a project template from a user terminal, generating project information, and providing a template corresponding to the selection result to the user terminal; And generating a field as desired by the user terminal, and receiving, by the user terminal, the field and the field value, which are finally inputted, as the field value is input.

Step (d) includes: (d-1) performing a refinement operation of removing errors due to logical contradictions between fields and field values or field values outside a preset range; And (d-2) removing fields and field values with low correlation by performing a correlation analysis between each field based on the medical data filtered through the purification operation.

Step (d-1) includes performing learning by referring to the medical data storage and calibration history of users previously stored in the server; Performing learning to analyze a condition for inputting a field value corresponding to an arbitrary field by referring to medical data from an external institution previously stored in the server; Setting a reference point for a numerical value of a field value corresponding to an arbitrary field by referring to a database for medical test result items and normal values of each item; And removing the items determined to be unnecessary based on the learning performed in step (d-1) and the reference point, and correcting the contradictory field values.

In step (d-2), statistical exposure frequency is analyzed by repetitively searching medical data having fields in common with fields remaining after refining in step (d-1), and within medical data having common fields. By performing a correlation analysis between the result value of and the remaining fields, fields and field values whose statistical significance is lower than a preset reference value are classified, and the classified fields and field values are removed.

Machine learning is performed through at least one of a convolutional neural network (CNN), a recurrent neural network (RNN), a long short-term memory (LSTM), and a natural language processing (NLP) algorithm.

(f) further comprising the step of bundling values having high correlation among the various numerical values constituting the output value provided to the user terminal, and converting the correlation relationship into a table or graph, and providing it.

In addition, the system for providing medical data collection and analysis service based on machine learning according to the second aspect of the present invention includes a server, and the server collects a plurality of medical data, performs pre-processing and classification on the collected data, and , A medical learning model is constructed by performing machine learning by referring to the classified medical data, and a user interface of a medical data collection and analysis platform is provided to the user terminal to randomize multiple medical data for the project to be analyzed by the user as desired. It is necessary to input multiple medical data inputted from the user terminal, remove or correct some data by performing optimization work on the plurality of medical data, and insert the collected medical data into the built medical learning model. The input and output values are provided to the user terminal, and the medical data includes medical-related institutional data including medical treatment data and clinical information data, and medical paper data.

The present invention provides a disease name, treatment method, and disease prognosis that are highly related to the project entered by the user simply by randomly inputting a plurality of medical data for a project to be analyzed by a user through a medical data collection and analysis platform user interface. An output value including prediction information and the like may be provided.

In addition, by using various medical-related data distributed and stored through the information management system for each medical information field, the medical data of the project entered by the user and the numerical values that are highly related to each other are grouped and the correlation between them is displayed in the form of a table or graph It can be provided by secondary processing.

1 is a block diagram of a system for providing a medical data collection and analysis service based on machine learning according to an embodiment of the present invention.
2 is a flowchart illustrating a method of collecting and analyzing medical data based on machine learning according to an embodiment of the present invention.
3 is a diagram for explaining a method of constructing a medical learning model by a server for providing medical data collection and analysis services based on machine learning according to an embodiment of the present invention.
4 is a diagram for explaining a method of optimizing medical data by a server for collecting and analyzing medical data based on machine learning according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those of ordinary skill in the art can easily implement the present invention. However, the present invention may be implemented in various different forms and is not limited to the embodiments described herein. In the drawings, parts irrelevant to the description are omitted in order to clearly describe the present invention, and similar reference numerals are attached to similar parts throughout the specification.

Throughout the specification, when a part is said to be "connected" with another part, this includes not only "directly connected" but also "electrically connected" with another element interposed therebetween. . In addition, when a part "includes" a certain component, it means that other components may be further included rather than excluding other components unless specifically stated to the contrary.

In the present specification, the term "unit" includes a unit realized by hardware, a unit realized by software, and a unit realized using both. Further, one unit may be realized by using two or more hardware, or two or more units may be realized by one piece of hardware. Meanwhile,'~ unit' is not meant to be limited to software or hardware, and'~ unit' may be configured to be in an addressable storage medium or configured to reproduce one or more processors. Thus, as an example,'~ unit' refers to components such as software components, object-oriented software components, class components, and task components, processes, functions, properties, and procedures. , Subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays and variables. Components and functions provided in the'~ units' may be combined into a smaller number of elements and'~ units', or may be further separated into additional elements and'~ units'. In addition, components and'~ units' may be implemented to play one or more CPUs in a device or a security multimedia card.

The "system" mentioned below may be implemented as a computer or portable terminal that can access a server or other terminal through a network. Here, the computer includes, for example, a notebook equipped with a web browser, a desktop, and a laptop.

The portable terminal is, for example, a wireless communication device that guarantees portability and mobility, and may be any kind of handheld-based wireless communication device such as, for example, a smart phone, a tablet PC, or a notebook.

In addition, a network refers to a connection structure in which information exchange between nodes such as terminals and servers is possible, and includes a local area network (LAN), a wide area network (WAN), and the Internet (WWW). : World Wide Web), wired and wireless data networks, telephone networks, wired and wireless television networks, etc. Examples of wireless data networks include 3G, 4G, 5G, 3GPP (3rd Generation Partnership Project), LTE (Long Term Evolution), WIMAX (World Interoperability for Microwave Access), Wi-Fi, Bluetooth communication, infrared communication, and ultrasound. Communication, Visible Light Communication (VLC), LiFi, and the like are included, but are not limited thereto.

1 is a block diagram of a system for providing a medical data collection and analysis service based on machine learning according to an embodiment of the present invention.

Referring to FIG. 1, a system for providing a medical data collection and analysis service based on machine learning according to an embodiment of the present invention includes a user terminal 100 and a medical data collection and analysis service providing server 200.

For example, the server 200 includes a communication module, a memory, a processor, and a database. The communication module processes data communication with the user terminal 100 and the information management system 300, respectively. A medical data collection and analysis service providing program for providing a medical data collection and analysis platform user interface to the user terminal 100 is stored in the memory. The processor executes the program stored in the memory, but performs processing corresponding to each operation of the medical data collection and analysis service providing program processed through the user terminal 100 to be described below.

Specifically, the user terminal 100 may be an intelligent terminal that is equipped with various functions, such as a smart phone, or that can freely mount and install various programs (eg, mobile apps, etc.).

The server 200 collects a plurality of medical data, performs pre-processing and classification on the collected data, and builds a medical learning model by performing machine learning by referring to the classified medical data. Provides a medical data collection and analysis platform user interface, allowing a user to randomly input a plurality of medical data for a project to be analyzed as desired, collect a plurality of medical data input from the user terminal 100, and Some data may be removed or corrected by performing an optimization operation on medical data, and an output value may be provided to the user terminal 100 by inputting a plurality of collected medical data into a built medical learning model. Here, the medical data includes medical-related institutional data and medical thesis data including medical treatment data and clinical information data.

For example, the server 200 may build a medical learning model by performing machine learning on a plurality of medical data collected from the information management system 300 constructed for each medical information field.

In addition, the user terminal 100 may create a project to be analyzed by the user through a medical data collection and analysis platform user interface. For example, a project may include a medical data input template for research that periodically investigates a patient's condition for a specific period of time, a hospital medical record template, a one-time medical information search template, a medical paper analysis template, a medical image learning template, etc. have.

Accordingly, the present invention uses various medical-related data distributed and stored through an information management system for each medical information field, and provides the name of the disease, treatment method and disease prognosis information, etc., which are highly correlated with the project to be analyzed by the user. You can provide an output value to include.

Hereinafter, a description of a configuration that performs the same function among the configurations illustrated in FIG. 1 will be omitted.

Hereinafter, a method of providing medical data collection and analysis service according to an embodiment of the present invention will be described in detail with reference to FIGS. 2 to 4.

2 is a flowchart illustrating a method of collecting and analyzing medical data based on machine learning according to an embodiment of the present invention.

Referring to FIG. 2, a method for providing medical data collection and analysis service based on machine learning includes a step of collecting a plurality of medical data, performing pre-processing and classification on the collected data (S100), with reference to the classified medical data. Step of constructing a medical learning model by performing machine learning (S200), by providing a medical data collection and analysis platform user interface to the user terminal 100 to randomly select a plurality of medical data for the project to be analyzed by the user. And collecting a plurality of medical data input from the user terminal 100 (S300), removing or correcting some data by performing an optimization operation on the plurality of medical data (S400), and a plurality of collected medical data And providing an output value to the user terminal 200 by inputting the medical data of the to the built medical learning model (S500). Here, the medical data includes medical-related institutional data including medical treatment data and clinical information data, and medical academic data including medical thesis data.

For example, in step S100, the server 200 may collect a plurality of medical data from the information management system 300 built for each medical information field. In addition, the server 200 may de-identify information corresponding to personal information among the collected medical data, and may perform labeling on the type of the collected medical data. For example, labeling can be performed to enable efficient identification and classification for each type of medical data, such as patient treatment data and clinical information data. Next, the labeled medical data can be classified by category.

3 is a view for explaining a method of constructing a medical learning model by a server for providing medical data collection and analysis service based on machine learning according to an embodiment of the present invention.

3, in step S200, the server 200 extracts keywords from medical data of each category, sets keywords related to symptoms or patient conditions as input values, and relates to disease results, treatment methods, or clinical results. Machine learning is performed by setting keywords as output values, and the machine learning algorithm is determined based on the category or type of medical data, and the optimal algorithm is selected through a fine tuning process that automatically performs a combination between a plurality of machine learning algorithms. Learning can be carried out.

As an example, machine learning for constructing a medical learning model by referring to classified medical data is a convolutional neural network (CNN), a recurrent neural network (RNN), a long short-term memory (LSTM), and a natural language processing (NLP). It can be performed through at least one of the algorithms.

In step S300, the server 200 may receive a selection result for a project template from the user terminal 100, generate project information, and provide a template corresponding to the selection result to the user terminal 100. In addition, the server 200 generates fields as desired by the user terminal 100 and inputs a field value, so that the user terminal 100 may receive the fields and field values that have been finally inputted.

For example, in step S300, the user terminal 100 may generate project information that the user wants to analyze through a medical data collection and analysis platform user interface provided from the server 200. For example, the user terminal 100 may generate project information by selecting a project template. For example, the project template may include a medical data input template, a hospital medical record template, a one-time medical information search template, a medical paper analysis template, and a medical image learning template for research that periodically investigates the patient's condition for a specific period of time. In addition, the user terminal 100 may select one of the template items.

In addition, after receiving the template selected by the user from the server 200, the user terminal 100 may generate a field suitable for the project characteristics according to a user input as desired. At this time, the field means the case report form. For example, the user terminal 100 is a desired field type suitable for the user's own project characteristics (e.g., disease symptoms, disease period, patient age, patient gender, patient residence, patient lifestyle, and pain). Information on the time point, intensity of pain, diagnosis (diagnosis code), representative symptoms, prescription name, etc.) and input method (e.g., table, text, image, number, check box, automatic completion using medical terminology library) You can create several and write the field value corresponding to the field. For example, when the field type is the gender of the patient, the field value means data about a man or woman, and when the field type is weight, the field value means an arbitrary value (eg, 50KG).

4 is a diagram for explaining a method of optimizing medical data by a server for collecting and analyzing medical data based on machine learning according to an embodiment of the present invention.

Referring to FIG. 4, the step of removing or correcting some data by performing an optimization operation on a plurality of medical data by the server 200 (S400) is an error due to a logical contradiction of a field and a field value or a preset range. Performing a refining operation for removing field values outside of (S410) and performing a correlation analysis between each field based on the medical data filtered through the refining operation to remove fields and field values with low correlation (S420) It may include.

Specifically, in step S410, the server 200 may perform learning by referring to the medical data storage and correction history of users previously stored in the database of the server 200. Subsequently, learning may be performed to analyze a condition for entering a field value corresponding to an arbitrary field by referring to medical data of an external institution previously stored in the server 200. For example, when the field value corresponding to the gender of the field type is'male', the'pregnant status' may be trained based on condition information that is unconditionally processed as'no'. Next, a reference point for a numerical value of a field value corresponding to an arbitrary field may be set by referring to the database for the medical test result items and the normal values of each item. Thereafter, the server 200 may remove items determined to be unnecessary based on the learning performed in step S410 and the reference point, and correct the contradictory field values. For example, in the case of collecting medical data related to pregnancy, the gender field, which is an unnecessary item, may be removed.

In step S420, the server 200 repeatedly searches for medical data having fields in common with the fields remaining after the refining operation to analyze the statistical exposure frequency, and the result values and remaining fields in the medical data having common fields. By performing a correlation analysis between the fields and field values having a statistical meaning lower than a preset reference value, the classified fields and field values may be removed. That is, finally, by extracting the refined data of meaningful fields, there is an effect of enabling effective learning with less data.

In addition, the server 200 may bundle values having a high correlation among various numerical values constituting the output value provided to the user terminal 100 and provide a table or a graph for the correlation between each other. For example, the output value may include disease name, treatment method, disease prognosis prediction information, and the like.

For example, the user terminal 100 may select an item to be analyzed and technique information to be applied from among a plurality of fields and field values. As an example, the server 200 refers to the item and technique information required to be analyzed received from the user terminal 100, and provides various information such as pValue hypothesis test, T test, network analysis, frequency analysis, Odds ratio, etc. It is possible to provide a set of statistics analyzed in real time. In this case, the statistics set means information visualized by graphs and tables, and for example, the relationship between alcohol consumption and dementia incidence rate among various treatment data can be automatically analyzed and provided without a separate setting.

As another example, the server 200 may perform an operation of further categorizing the collected data. For example, the user terminal 100 may print in a report form or provide and share an extracted pdf file or web page in various ways.

An embodiment of the present invention may also be implemented in the form of a recording medium including instructions executable by a computer, such as a program module executed by a computer. Computer-readable media can be any available media that can be accessed by a computer, and includes both volatile and nonvolatile media, removable and non-removable media. Further, the computer-readable medium may include a computer storage medium. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data.

Although the methods and systems of the present invention have been described in connection with specific embodiments, some or all of their components or operations may be implemented using a computer system having a general-purpose hardware architecture.

The above description of the present invention is for illustrative purposes only, and those of ordinary skill in the art to which the present invention pertains will be able to understand that other specific forms can be easily modified without changing the technical spirit or essential features of the present invention. will be. Therefore, it should be understood that the embodiments described above are illustrative and non-limiting in all respects. For example, each component described as a single type may be implemented in a distributed manner, and similarly, components described as being distributed may also be implemented in a combined form.

The scope of the present invention is indicated by the claims to be described later rather than the detailed description, and all changes or modified forms derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present invention do.

100: user terminal
200: server
300: information management system

Claims (10)

In the method of providing medical data collection and analysis services based on machine learning,
(a) collecting a plurality of medical data, and performing pre-processing and classification on the collected data;
(b) constructing a medical learning model by performing machine learning by referring to the classified medical data;
(c) Provides a medical data collection and analysis platform user interface to the user terminal, allowing the user to randomly input multiple medical data for the project to be analyzed as desired, and to collect a plurality of medical data input from the user terminal. step;
(d) removing or correcting some data by performing an optimization operation on the plurality of medical data; And
(e) inputting the collected plurality of medical data to the constructed medical learning model and providing an output value to the user terminal,
The medical data includes medical-related institutional data including medical treatment data and clinical information data, and medical academic data including medical thesis data. A method of collecting and analyzing medical data based on machine learning.
The method of claim 1,
The step (a),
Performing de-identification processing of information corresponding to personal information among the collected data to be removed from the collection object or processed so that humans do not recognize it;
Labeling the types of the collected data; And
A method for providing a medical data collection and analysis service based on machine learning, comprising the step of classifying the labeled data by category.
The method of claim 2,
The step (b),
Machine learning is performed by extracting keywords from the medical data of each category, setting keywords related to symptoms or patient conditions as input values, and setting keywords related to disease results, treatment methods, or clinical results as output values. Machine learning is performed by determining a machine learning algorithm based on the category or shape of and selecting the optimal algorithm through a fine tuning process that automatically performs a combination between a plurality of machine learning algorithms.
A method of providing medical data collection and analysis services based on machine learning.
The method of claim 1,
The step (c),
Receiving a selection result for a project template from the user terminal, generating project information, and providing a template corresponding to the selection result to the user terminal; And
Generating a field as desired in the user terminal, and receiving a field and field value finally inputted by the user terminal as the field value is input; including,
A method of providing medical data collection and analysis services based on machine learning.
The method of claim 4,
The step (d),
(d-1) performing a refinement operation of removing errors due to logical contradictions of the field and field values or field values outside a preset range; And
(d-2) removing fields and field values with low correlation by performing correlation analysis between fields based on the medical data filtered through the refining operation; including,
A method of providing medical data collection and analysis services based on machine learning.
The method of claim 5,
The step (d-1),
Performing learning by referring to the medical data storage and calibration history of users previously stored in the server;
Performing learning to analyze a condition for inputting a field value corresponding to an arbitrary field by referring to medical data from an external institution previously stored in the server;
Setting a reference point for a numerical value of a field value corresponding to an arbitrary field by referring to a database for medical test result items and normal values of each item; And
A method for providing medical data collection and analysis service based on machine learning, comprising removing items determined to be unnecessary and correcting contradictory field values based on the learning performed in step (d-1) and the reference point.
The method of claim 6,
The step (d-2),
In step (d-1), medical data having a field remaining after the refining operation and a common field are repeatedly searched to analyze the statistical exposure frequency, and the result value in the medical data having the common field and the remaining A method for providing medical data collection and analysis service based on machine learning, which is to perform correlation analysis between fields, classify fields and field values whose statistical significance is lower than a preset reference value, and remove the classified fields and field values.
The method of claim 1,
The machine learning is performed through at least one of CNN (Convolutional Neural Network), RNN (Recurrent Neural Nerwork), LSTM (Long Short-Term Memory), and NLP (Natiral Language Processing) algorithm. How to provide data collection and analysis services.
The method of claim 1,
(f) Collecting medical data based on machine learning, further comprising the step of grouping values having high correlation among various numerical values constituting the output value provided to the user terminal and converting the correlation relationship to a table or graph, and providing How to provide analytics services.
In a system for providing medical data collection and analysis services based on machine learning,
Includes a server,
The server,
Collect multiple medical data, perform pre-processing and classification on the collected data,
Constructing a medical learning model by performing machine learning by referring to the classified medical data,
Provides a medical data collection and analysis platform user interface to the user terminal, allowing the user to randomly input multiple medical data for the project to be analyzed as desired, and collects a plurality of medical data input from the user terminal,
Some data is removed or corrected by performing an optimization operation on the plurality of medical data,
Inputting the collected plurality of medical data to the constructed medical learning model and providing an output value to the user terminal,
The medical data includes medical-related institutional data including medical treatment data and clinical information data, and medical academic data including medical thesis data. A system for providing medical data collection and analysis service based on machine learning.

Images