KR20210075028A - System and Method for providing individual health-care information from AI database to user's device - Google Patents

Abstract

Proposed, in the present invention, are a system and method for providing individual health-care information in real-time from an artificial intelligence database to a user terminal device. The artificial intelligence database provides the health-care information necessary for the user in real-time based on a user health measurement data obtained from the terminal device or a device for which the user uses in his/her daily life. Therefore, the present invention is capable of having an effect of preserving health in advance.

Description

System and Method for providing individual health-care information from AI database to user's device in real time from an artificial intelligence database server to a user terminal

The present invention relates to a system and method for providing individualized management information to a user.

The traditional method of collecting health information in medicine is to record disease-related information in an Electronic Health Record when a patient visits a medical institution such as a hospital or clinic. Although this method enables medical institutions to intensively grasp information about patients and provide treatment, there is a limitation in that necessary health information for patients needs to be collected within a short period of time. Therefore, in addition to the problem of the patient's current health status, it is difficult to collect daily lifestyle information about daily life such as the patient's lifestyle, habits, and surrounding environment, which are the root causes of the patient's health status, and even such daily lifestyle information Even if data is collected, it has the disadvantage that it may be inaccurate because it depends on the patient's subjective memory. In addition, there is a burden that a considerable amount of time must be spent to obtain information about the patient's daily lifestyle. In addition, since the traditional method investigates disease-related information on patients who have visited medical institutions, it is difficult to collect information about daily lifestyles for healthy people before specific symptoms or diseases appear or for people who do not have awareness of the disease. There are limits.

Recently, with the development of the Internet, wearable devices, and smartphones, it is possible to monitor individual bio-signals every day, and it is possible to detect changes in bio-signals and abnormal signs occurring in individuals at an early stage. However, even when individual bio-signals are collected in real time, biometric information collected only through wearable devices and smart phones is limited in understanding the health status of users.

KR 10-1808541 B1

In a preferred embodiment of the present invention, based on the user's health data acquired and collected from the user terminal, the user detects an abnormality in health and provides individualized health management information to each user even before visiting the hospital.

In a preferred embodiment of the present invention, by using the user's health data obtained and collected from the user terminal, public health examination data, and artificial intelligence, when an abnormality occurs in the user's health, problems may occur to the user at an early stage. As a reminder, we want to provide a health management information platform that allows users to recommend treatment or activities before they suffer permanent damage.

As a preferred embodiment of the present invention, a system for providing individualized health management information from an artificial intelligence database server to a user terminal in real time, the system comprising an artificial intelligence database server and a user terminal, wherein the user terminal includes at least one A machine learning unit that communicates with the wearable device of the AI database server and performs principal component analysis training on the learning data stored as a matrix table by region, gender, year, and month in a public health examination data database; The relationship between each of the first principal component and at least one or more second principal components analyzed through the principal component analysis training in the machine learning unit, and the amount of change of the first principal component and the amount of change of each of the at least one or more second principal components is two-dimensional. A display unit for displaying a health checkup factor analysis result as a dot graph of ; and performing random forest learning based on the correlation between the first main component displayed as the dot graph and the at least one second main component to be related to the first main component It is characterized in that it includes; a recommendation unit for recommending health management information.

As a preferred embodiment of the present invention, in a system for providing individualized health management information in real time from an artificial intelligence database server to a user terminal, the user terminal is at least one of the user's at least one wearable device used by the user in real time. an individual health information acquisition unit that secures one health measurement data; and a health management unit that searches for the first main component corresponding to the at least one health measurement data of the user, and receives and provides health management information related to the first main component from the recommendation unit.

As a preferred embodiment of the present invention, in a system for providing individualized health management information in real time from an artificial intelligence database server to a user terminal, the recommendation unit provides the user terminal with health measurement data information measured or collected by the user terminal, the Information on whether the health measurement data falls within the normal range compared to the health reference table, health management information related to the first principal component learned through principal component analysis training analysis in the machine learning unit based on health data corresponding to the health measurement data , and characterized by providing health management information related to the first principal component learned by performing random forest learning in the recommendation unit.

As a preferred embodiment of the present invention, in a system for providing individualized health management information from an artificial intelligence database server to a user terminal in real time, the public health examination data database includes health examination information and treatment history information opened by the National Health Insurance Corporation. and processing and storing drug prescription information learning data.

As a preferred embodiment of the present invention, in a system for providing individualized health management information from an artificial intelligence database server to a user terminal in real time, the recommendation unit is the health required by the expert for the first main component or the amount of change in the first main component When the contents of the management information are written, the random forest learning is performed based on the contents of the prepared health management information.

As another preferred embodiment of the present invention, the method of providing individualized health management information in real time from the artificial intelligence database server to the user terminal is the learning stored as a matrix table by region, gender, year, and month in the public health examination data database. performing principal component analysis training on the data through a machine learning unit; The relationship between the first principal component and the at least one or more second principal components analyzed through the principal component analysis training in the display unit, or the relationship between the amount of change of the first principal component and the amount of change of each of the at least one or more second principal components is expressed as a two-dimensional dot graph. displaying; and recommending health care information related to the first principal component by performing random forest learning based on the correlation between the first principal component and the at least one or more second principal components displayed in the dot graph by the recommendation unit. characterized in that

As a preferred embodiment of the present invention, in the method of providing individualized health management information in real time from an artificial intelligence database server to a user terminal, the artificial intelligence database server performs wired/wireless communication with the user terminal, and the user terminal securing at least one health measurement data of the user acquired in real time from at least one wearable device used by the user; and receiving the health measurement data secured from the user terminal by the recommendation unit and determining whether there is health measurement data deviating from the health standard table stored in the artificial intelligence database server; It is characterized in that it further comprises.

As a preferred embodiment of the present invention, in a method of providing individualized health management information from an artificial intelligence database server to a user terminal in real time, the recommendation unit responds to the health measurement data when there is health measurement data that deviate from the health standard table retrieving the first principal component, and recommending health management information related to the first principal component to the user terminal by performing random forest learning based on the correlation between the at least one or more second principal components. characterized in that

As a preferred embodiment of the present invention, in a method of providing individualized health management information in real time from an artificial intelligence database server to a user terminal, the recommendation unit provides the user terminal with health measurement data information measured or collected by the user terminal, the Information on whether the health measurement data falls within the normal range compared to the health reference table, health management information related to the first principal component learned through principal component analysis training analysis in the machine learning unit based on health data corresponding to the health measurement data , and characterized by providing health management information related to the first principal component learned by performing random forest learning in the recommendation unit.

As a preferred embodiment of the present invention, a system and method for providing individualized health management information in real time to a user terminal in an artificial intelligence database is based on the user's health measurement data obtained from the terminal or device that the user uses in daily life. By providing necessary health management information to users in real time, it has the effect of detecting abnormalities in the user's health at an early stage and protecting their health in advance.

1 shows an internal configuration diagram of a system that provides individualized health management information in real time to a user terminal in an artificial intelligence database as a preferred embodiment of the present invention.
2 to 4 are, as a preferred embodiment of the present invention, reprocessing and storing health data stored in the Health Insurance Corporation database in the form of a matrix table by region, gender, year, and month in order to learn the learning data in the machine learning unit An example is shown.
5 to 6 show an example of a dot graph related to triglyceride for men aged 45 to 49 in Seoul in the display unit as a preferred embodiment of the present invention.
7 shows an example of a health standard table pre-stored in an artificial intelligence database as a preferred embodiment of the present invention.
8 shows an example of providing individualized health management information to a user terminal as a preferred embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS [0010] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS [0010] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS [0023] Reference is made to the accompanying drawings, which show by way of illustration specific embodiments in which the invention may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the present invention. It should be understood that the various embodiments of the present invention are different but need not be mutually exclusive. For example, certain shapes, structures, and characteristics described herein may be implemented with changes from one embodiment to another without departing from the spirit and scope of the present invention. In addition, it should be understood that the location or arrangement of individual components within each embodiment may be changed without departing from the spirit and scope of the present invention. Accordingly, the following detailed description is not to be taken in a limiting sense, and the scope of the present invention should be taken as encompassing the scope of the claims and all equivalents thereto. In the drawings, like reference numerals refer to the same or similar elements throughout the various aspects.

Hereinafter, in order to enable those of ordinary skill in the art to easily practice the present invention, various embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a diagram showing the internal configuration of a system that provides individualized health management information in real time to a user terminal in an artificial intelligence database as a preferred embodiment of the present invention.

The system 100 shown in FIG. 1 includes an artificial intelligence database server 120 and a user terminal 110 . At this time, the user terminal 110 performs communication with at least one wearable device, other various medical devices and terminals, and acquires health data including biometric data such as blood pressure, pulse, body temperature, and blood sugar of the user in real time. and accumulated, collected and managed.

The user terminal 110, at least one wearable device, and various other medical devices and terminals are UE (User Equipment), ME (Mobile Equipment), MS (Mobile Station), UT (User Terminal), SS (Subscriber Station), MSS(Mobile Subscriber Station), wireless device, handheld device, AT (Access Terminal), wearable device, robot, etc. are all concepts including mobile phone, cellular phone, smart phone, It can be implemented in the form of PDA (Personal Digital Assistants), PMP (Portable Multimedia Player), tablet device, computer, smart watch, CCTV, artificial intelligence speaker, home appliance or multimedia device.

As another embodiment of the present invention, the terminal 110 includes a camera, a display, a memory and a communication unit, and may have an interface for transmitting and receiving image data or audio data. The user terminal 110 may also download and install an application that automatically detects a user's health abnormality in the form of an app or provides health management information suitable for the user's health condition from the artificial intelligence database 120 .

As a preferred embodiment of the present invention, the artificial intelligence database server 120 that provides individualized health management information to each user is capable of wired/wireless communication with the Health Insurance Corporation database server 160 and the like. In addition, when an emergency occurs, it is possible to notify the user of the emergency situation through a call center server or a hospital server.

As a preferred embodiment of the present invention, the artificial intelligence database server 120 that provides individualized health management information to each user includes a machine learning unit 130 , a display unit 140 , and a recommendation unit 150 .

The machine learning unit 130 performs principal component analysis training on the learning data stored as a matrix table by region, gender, year, and month in a public health examination data database such as the Health Insurance Corporation database 160 . Principal component analysis refers to a machine learning method that finds the axis that maximizes the variance of learning data composed of various factors (dimensions).

In a preferred embodiment of the present invention, the machine learning unit 130 finds the first axis with the largest variance in the learning data obtained from the National Health Insurance Corporation database 160, and finds the first axis with the largest variance while being orthogonal to the found first axis. Find the second axis. Afterwards, the third axis with the maximum variance is found while being orthogonal to the first and second axes. The machine learning unit 130 finds the axis as many as the target dimension while repeating the step of finding the third axis from the first.

In a preferred embodiment of the present invention, the machine learning unit 130 intends to use the health data of the National Health Insurance Corporation database 160 in which medical treatment history information, drug prescription information, and health checkup information are accumulated. The machine learning unit 130 generates learning data by processing and storing health data in which medical treatment history information, drug prescription information, and health check-up information of national health insurance subscribers are accumulated in the format shown in FIGS. Machine learning is performed using data.

The National Health Insurance Corporation uses the National Health Insurance Corporation database 160 to provide the subscriber serial number and basic information (gender, age, and residence code), body size information such as body, weight, and waist circumference, blood pressure, blood sugar, cholesterol, and urine protein. , results of pathological tests such as gamma GPTI, results of diagnostic tests such as visual acuity, hearing, and oral examinations, as well as the results of other questionnaires on whether to drink or smoke, etc.

Referring to FIG. 1 , the Health Insurance Corporation database 160 includes a health checkup information storage unit 162 , a treatment history information storage unit 164 , and a medicine prescription information storage unit 166 .

FIG. 2 shows an example in which the machine learning unit 130 processes data stored in the health checkup information storage unit 162 as a preferred embodiment of the present invention. The machine learning unit 130 compares the data stored in the health checkup information storage unit 162 by referring to "the standard for determining the results of general health checkup and medical benefit life-time transition period checkup" for reference year, province, gender, age group (5-year-old unit), Learning is performed to derive correlations with the examination results based on height (in units of 5 cm) and weight (units of 5 kg).

FIG. 3 shows an example in which the machine learning unit 130 processes data stored in the treatment history information storage unit 164 as a preferred embodiment of the present invention.

The machine learning unit 130 classifies the data stored in the treatment history information storage unit 164 into the base year, province, gender, and age group with reference to the "disease classification code", and thereafter, the treatment subject code, the common disease code, and the injured disease code. , the number of days of treatment, the number of days in the hospital, etc., the learning is performed to derive the correlation of the examination result.

4 shows an example in which the machine learning unit 130 processes data stored in the medicine prescription information storage unit 166 as a preferred embodiment of the present invention.

The machine learning unit 130 also classifies the data stored in the drug prescription information storage unit 166 into base year, province, gender, and age group with reference to the "Ministry of Health and Welfare Announced Drug Reimbursement List and Salary Upper Amount Table", and classifies the drug generic name code , and the total number of days of administration, learning is carried out to derive correlations with the examination results.

As a preferred embodiment of the present invention, the machine learning unit 130 performs Principal Component Analysis training for each matrix table by region, gender, year, and month on the training data shown in FIGS. 2 to 4 . Then, the principal component analysis training analysis result is provided to the display unit 140 .

5 to 6 show an example of a dot graph related to triglyceride for men aged 45 to 49 in Seoul in the display unit as a preferred embodiment of the present invention.

The display unit 140 displays the relationship between the first principal component and at least one or more second principal components analyzed through principal component analysis training in the machine learning unit 130, the amount of change in the first principal component, and the amount of change in each of the at least one or more second principal components. The relationship between the two is displayed as a two-dimensional dot graph to display the result of health checkup factor analysis. In FIG. 5 , the first main component is triglyceride, and the second main component is gamma_gtp and idl_chole.

Referring to FIG. 6 , through the principal component analysis training analysis in the machine learning unit, triglyceride as the first main component and other components related to triglyceride include weight, waist, Cholesterol (tot_chole, ldl_chole, hdl_chole), liver function tests (SGPT(ALT), SGOT(AST), gamma_gtp), kidney, creatine, blood pressure (bp_lwst, bp_high, bids) were derived. And, gamma_gtp and ldl_chole were derived as components highly related to triglycerides in men aged 45 to 49 in Seoul.

The recommendation unit 150 performs random forest learning based on the relationship between the first principal component displayed as a graph on the display unit 140 and at least one or more second principal components. If the expert writes the health management information required for the first main component or the amount of change of the first main component before performing the random forest learning, the recommendation unit 150 performs random forest learning based on the prepared health management information. can When the random forest learning is completed, the recommendation unit 150 recommends health management information related to the first main component.

For example, the recommendation unit 150 is based on the correlation between the first main component triglyceride and the second main components gamma_gtp and ldl_chole, which are identified in the dot graph displayed on the display unit 140 as shown in FIGS. 4 to 5 . Random forest learning is performed. Then, based on the random forest learning result, health management information related to the first main component is recommended.

For example, when the triglyceride measurement sensor installed in the user terminal 110 acquires health data of triglyceride 173 of a 47-year-old man living in Seoul, the recommendation unit 150 sends a message to the city of Seoul received from the user terminal 110 . The company searches the artificial intelligence database server 120 for the first main component corresponding to the triglyceride 173 value of a 47-year-old male. When there is data on the first main component in the artificial intelligence database server 120 , health management information related to the first main component is transmitted to the user terminal 110 . When the data learned about triglycerides is stored and managed in the artificial intelligence database server 120 as shown in FIGS. 5 to 6 , the recommendation unit 150 is the health of FIG. 7 stored in the artificial intelligence database server 120 in advance. Referring to the same table as the reference table, as in the example of FIG. 8 , the triglyceride in the user terminal 110 is 173 and belongs to the normal B (boundary) 810, and if it becomes higher, 'dyslipidemia' may occur, Numerical testing of gamma_gtp and ldl_chole is necessary, and it may be suggested to abstain from alcohol and tobacco, and to exercise regularly.

As a preferred embodiment of the present invention, the recommendation unit 150 determines whether the health measurement data information 811 measured by the user terminal and the health measurement data of the user fall within the normal range compared to the health standard table (refer to FIG. 7) in the user terminal. Health management information (830, 840) related to the first principal component learned through principal component analysis training analysis in the machine learning unit 130 on the basis of the information on whether or not information (810, 820) and health data corresponding to the user's health measurement data ), the recommendation unit 150 may provide the health management information 850 related to the learned first principal component by performing random forest learning. The health measurement data information 811 includes, in addition to the user's blood pressure, pulse, heart rate, and other types of health-related data that can be measured through the terminal, the region in which the user resides, the user's gender, and the year in which the data was measured. Contains information about dates such as (year) and month (month).

Methods according to an embodiment of the present invention may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc. alone or in combination. The program instructions recorded on the medium may be specially designed and configured for the present invention, or may be known and available to those skilled in the art of computer software.

As described above, although the present invention has been described with reference to limited embodiments and drawings, the present invention is not limited to the above embodiments, and various modifications and variations from these descriptions are provided by those skilled in the art to which the present invention pertains. This is possible.

Claims (10)

A system that provides individualized health management information in real time from an artificial intelligence database server to a user terminal, the system comprising:
It includes an artificial intelligence database server and a user terminal, wherein the user terminal performs communication with at least one wearable device,
The artificial intelligence database server is
a machine learning unit that performs principal component analysis training on the learning data stored as a matrix table by region, gender, year and month in the public health examination data database;
The relationship between each of the first principal component and at least one or more second principal components analyzed through the principal component analysis training in the machine learning unit, and the amount of change of the first principal component and the amount of change of each of the at least one or more second principal components is two-dimensional. A display unit that displays the health checkup factor analysis result as a dot graph of; And
and a recommendation unit for recommending health care information related to the first principal component by performing random forest learning based on the correlation between the first principal component and the at least one or more second principal components represented by the dot graph. system that does. The method of claim 1, wherein the user terminal is
an individual health information acquisition unit for securing at least one health measurement data of the user acquired in real time from at least one wearable device used by the user; and
a health management unit that searches for the first main component corresponding to the at least one health measurement data of the user, and receives and provides health management information related to the first main component from the recommendation unit;
The health measurement data includes, in addition to the user's blood pressure, pulse, heart rate, and other health-related data that can be measured through the user terminal, the region in which the user resides, the user's gender, the year in which the data was measured, A system comprising information about a date, such as a month. The method of claim 2, wherein the recommendation unit
The machine learning based on health measurement data information measured or collected by the user terminal in the user terminal, information on whether the health measurement data falls within a normal range compared to a health reference table, and health data corresponding to the health measurement data A system characterized in that it provides health management information related to the first principal component learned through principal component analysis training analysis in the unit, and health management information related to the first principal component learned by performing random forest learning in the recommendation unit. The method of claim 1, wherein the public health examination data database is
A system characterized by processing and storing health checkup information, treatment history information, and drug prescription information learning data opened by the National Health Insurance Corporation. The method of claim 1, wherein the recommendation unit
When the expert writes the health management information required for the first main component or the amount of change of the first main component, the random forest learning is performed based on the prepared health management information. A method of providing individualized health management information in real time from an artificial intelligence database server to a user terminal, comprising:
performing principal component analysis training through a machine learning unit on the learning data stored as a matrix table by region, gender, year, and month in a public health examination data database;
The relationship between the first principal component and at least one or more second principal components analyzed through the principal component analysis training in the display unit, or the relationship between the amount of change of the first principal component and the amount of change of each of the at least one or more second principal components, is expressed as a two-dimensional dot graph. displaying; and
Recommending health management information related to the first principal component by performing random forest learning on the basis of the correlation between the first principal component and the at least one or more second principal components displayed in the dot graph in the recommendation unit; How to characterize. 7. The method of claim 6,
The artificial intelligence database server performs wired/wireless communication with a user terminal, and the user terminal secures at least one health measurement data of the user obtained in real time from at least one wearable device used by the user; and
receiving the health measurement data secured from the user terminal by the recommendation unit and determining whether there is health measurement data that deviates from the health standard table stored in the artificial intelligence database server; further comprising,
The health measurement data includes, in addition to the user's blood pressure, pulse, heart rate, and other health-related data that can be measured through the user terminal, the region in which the user resides, the user's gender, the year in which the data was measured, A method comprising information about a date, such as a month. 8. The method of claim 7,
When there is health measurement data that deviates from the health standard table, the recommendation unit searches for the first principal component corresponding to the corresponding health measurement data, and performs random forest learning based on the correlation between the at least one or more second principal components. The method of claim 1, further comprising: recommending health management information related to a first main ingredient to the user terminal. The method of claim 1, wherein the public health examination data database is
It processes and stores the health examination information, treatment history information, and drug prescription information learning data opened by the National Health Insurance Corporation, and the machine learning unit performs learning by using the learning data processed and stored in the public health examination data database as big data. A method characterized in that it is performed. The method of claim 8, wherein the recommendation unit
The machine learning based on health measurement data information measured or collected by the user terminal in the user terminal, information on whether the health measurement data falls within a normal range compared to a health reference table, and health data corresponding to the health measurement data A method for providing health management information related to the first principal component learned through principal component analysis training analysis in the unit, and health management information related to the first principal component learned by performing random forest learning in the recommendation unit.

Images