KR102216236B1 - Method for providing medical information related to health contents based on AI(artificial intelligence) and apparatus for performing the method - Google Patents

Abstract

The present invention relates to a method for providing medical information related to health content based on artificial intelligence, and to a device for performing the method. The method for providing medical information related to health content based on artificial intelligence comprises the steps of: determining health content by a medical information providing service server; determining medical information related to the health content by the medical information providing service server; and transmitting the medical information to a medical information receiving server by the medical information providing service server.

Description

TECHNICAL FIELD [Method for providing medical information related to health contents based on AI (artificial intelligence) and apparatus for performing the method]

The present invention relates to a method for providing medical information related to health content based on artificial intelligence and an apparatus for performing the method. More specifically, the present invention relates to a method of providing medical information related to health content by analyzing health content existing in a medium such as a web page, and an apparatus for performing the method.

Recently, medical consumers are discussing the right way to utilize online health content. A large percentage of the population uses the Internet, and a large percentage of these Internet users search for or use health information through health contents provided through the Internet.

In particular, people constantly search online for information shared by people who have experienced the same symptoms and situations as themselves, and check their health status through health contents that exist on the Internet.

In terms of time and money, the Internet is highly efficient, and while acquiring health content through traditional health information media such as TV, newspapers, and medical experts and medical journals is effective, it is only providing health information. In order to receive additional information on additional treatment or solutions, additional search is required.

Therefore, there is a need for research on a method for directly providing additional information related to health contents based on an algorithm.
Prior art is disclosed in Korean Patent Application Publication No. 10-2019-0076381 (2019.07.02.).

An object of the present invention is to solve all of the above-described problems.

In addition, an object of the present invention is to simply provide medical information related to health content to a user without additional search by the user based on additional provision of medical information related to health content based on artificial intelligence.

In addition, the present invention provides information on hospital information, pharmacy information, and health food through tools such as a widget on a website, so that it is simply installed on any site to separately provide medical information related to health content. It is aimed at.

A typical configuration of the present invention for achieving the above object is as follows.

According to an embodiment of the present invention, a method of providing medical information related to health content based on artificial intelligence comprises the steps of: a medical information providing service server determining health content, and the medical information providing service server determining medical information related to the health content. And transmitting, by the medical information providing service server, the medical information to a medical information receiving server.

Meanwhile, the medical information receiving server is a server that provides the health content, and the medical information receiving server may provide the medical information related to the health content together with the health content.

In addition, the determining of the health content includes the steps of extracting, by the medical information receiving server, a health keyword from the input content, and determining whether the input content is the health content based on the health keyword. It may include steps.

According to another embodiment of the present invention, a medical information providing service server that provides medical information related to health content based on artificial intelligence includes a communication unit implemented to receive input contents and a processor operatively connected to the communication unit. However, the processor may be implemented to determine health content among the input content, determine medical information related to the health content, and transmit the medical information to a medical information receiving server.

Meanwhile, the medical information receiving server is a server that provides the health content, and the medical information receiving server may provide the medical information related to the health content together with the health content.

Further, the processor may be implemented to extract a health keyword from the input content and determine whether the input content is the health content based on the health keyword.

According to the present invention, based on the additional provision of medical information related to health content based on artificial intelligence, a user can simply obtain medical information related to health content without additional search by the user.

In addition, according to the present invention, by providing information on hospital information, pharmacy information, and health food through tools such as widgets on a web site, medical information related to health content is separately provided by being simply installed on any site. Can be.

1 is a conceptual diagram showing a system for providing medical information related to health content based on artificial intelligence according to an embodiment of the present invention.
2 is a conceptual diagram illustrating an operation of a medical information providing service server according to an embodiment of the present invention.
3 is a conceptual diagram showing an operation of a content reliability adjustment unit according to an embodiment of the present invention.
4 is a conceptual diagram showing a method for matching health content and medical information according to an embodiment of the present invention.
5 is a conceptual diagram showing a method of providing medical information according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE INVENTION The detailed description of the present invention to be described later refers to the accompanying drawings, which illustrate specific embodiments in which the present invention may be practiced. These embodiments are described in detail sufficient to enable those skilled in the art to practice the present invention. It is to be understood that the various embodiments of the present invention are different from each other, but need not be mutually exclusive. For example, specific shapes, structures, and characteristics described herein may be changed from one embodiment to another and implemented without departing from the spirit and scope of the present invention. In addition, it should be understood that the positions or arrangements of individual elements in each embodiment may be changed without departing from the spirit and scope of the present invention. Therefore, the detailed description to be described later is not intended to be limited, and the scope of the present invention should be taken as encompassing the scope claimed by the claims of the claims and all scopes equivalent thereto. Like reference numerals in the drawings indicate the same or similar elements over several aspects.

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

1 is a conceptual diagram illustrating a system for providing medical information related to health content based on artificial intelligence according to an embodiment of the present invention.

In FIG. 1, a system for providing medical information related to health content for providing medical information related to health content based on artificial intelligence is disclosed.

Referring to FIG. 1, a system for providing medical information related to health content may include a medical information providing service server 100, a medical information receiving server 150, and a user device 170.

The medical information providing service server 100 may include a health content determination unit 110, a medical information determination unit 120, and a machine learning unit 130.

The health content determination unit 110 may be connected to the machine learning unit 130 to collect content existing on a medium such as a web page, and may be implemented to determine whether the collected content is health content. The health content determination unit 110 may collect health content existing on a web page set for a job such as web crawling or a pre-search.

Health content collected by the health content determination unit 110 may be classified according to purpose. Specifically, the health content collected by the health content determination unit 110 is a health content for learning by the machine learning unit 130, and a target for providing medical information related to health content other than the learning by the machine learning unit 130 It may include health content that becomes.

The medical information determination unit 120 may be implemented to provide medical information related to the health content when the input content delivered by the health content determination unit 110 is determined as health content. The medical information may include hospital information, pharmacy information, drug information, and the like related to health content.

The medical information determination unit 120 may be connected to the machine learning unit 130 to determine medical information related to health content. The medical information determination unit 120 may transmit health content to the machine learning unit 130, and the machine learning unit 130 may provide medical information related to the health content based on the machine learning result. Alternatively, the medical information determining unit 120 may determine medical information related to health content based on the learning model learned by the machine learning unit 130 and the learning model.

The medical information providing service server 100 may transmit medical information related to the health content determined by the medical information determining unit 120 to the medical information receiving server 150.

The machine learning unit 130 may be implemented for machine learning to determine whether the input content is health content, and to determine medical information related to the health content when the health content is input. The machine learning unit 130 may determine health content and perform machine learning on a relationship between health content and medical information in various ways. A specific learning method of the machine learning unit 130 will be described later.

The medical information receiving server 150 may receive medical information related to health content from the medical information providing service server 100 and provide the related medical information on a web page on the health content. For example, the medical information receiving server 150 may provide medical information related to health content on a medical providing module such as a widget or a pop-up on a web page providing health content. A medical providing module may be provided on the user device through the medical information receiving server 150 by a simple operation such as inserting a script into a web site.

The user device 170 may receive medical information added on the medical information receiving server 150 together with health content. When the user device 170 receives a web page related to health content, medical information related to the health content may be provided together with the web page related to the health content on the user device.

2 is a conceptual diagram illustrating an operation of a medical information providing service server according to an embodiment of the present invention.

In FIG. 2, a method for a medical information providing service server to provide medical information related to health content is disclosed. In particular, a method for a medical information providing service server to determine whether content is health content and extracting a health keyword from the determined health content is disclosed.

Referring to FIG. 2, the health keyword is a word related to health and may be used to determine medical information related to health content.

The health content determination unit 210 may include a primary health content determination model 230 learned based on the primary health content 220. The primary health content 220 is previously collected health content and may be content used for learning by the health content determination unit 210. The health content determination unit 210 may determine whether the input content is the health content (discrimination) 260 based on the model learned based on the primary health content 220. The primary health content determination model 230 may learn about health keywords based on the frequency of words used in health content and determine whether the input content is health content based on the health keywords.

The primary health content determination model 230 extracts health keywords from the primary health content 220, and based on each of the extracted health keywords, the first health content reliability 240 for the input content, and the input content Learning to determine a second health content reliability 250 for a plurality of health keyword combinations may be performed.

For example, if the input content includes {chills, fever, cough corona} as the health keyword, the first health content reliability 240 is based on each of {chills}, {fever}, and {cough} {corona} It may be the reliability of whether or not the input content determined as health content, and the second health content reliability 250 is health keywords such as {chills, fever}, {cough, corona}, {chills, fever, cough corona} It may be the reliability of whether the input content determined based on each combination is health content.

The first health content reliability 240 for each of the plurality of health keywords is a probability that the input content is health content when the input content includes each of the plurality of health keywords. The first health content reliability 240 may be individually determined for each of the plurality of health keywords. The second health content reliability 250 for the health keyword combination is a probability that the input content is health content when the input content includes a specific health keyword combination. Reliability may be determined based on the same/similarity between keywords included in the health content and input content keywords.

For example, when one input content is input, a plurality of health keywords included in the input content may be extracted, and each of the plurality of first health content reliability 240 for each of the plurality of health keywords is determined, Each of the plurality of second health content reliability levels 250 for the health keyword combination may be determined.

The first health content reliability 240 and the second health content reliability 250 may be determined based on weights set for each health keyword. The weight of the health keyword may have a higher value as the health keyword is a medical term. For example, the probability that content containing the term'ankylosing spondylitis' is health content is higher than that of the liver content containing the term'cough' is health content. Therefore, in an embodiment of the present invention, classification of a terminology may be performed by learning a separate terminology in the primary health content determination model 230, and a weight for the terminology may be set to be relatively higher, Reliability can be determined. Additionally, in the primary health content determination model 230, the lower the frequency of use in the primary health content 220 is set to have a higher weight in consideration of the frequency of use in the primary health content 220 among the specialized medical terms. I can.

When the maximum value of the plurality of first health content reliability 240 is greater than or equal to the first threshold value, and the maximum value of the plurality of second health content reliability level 250 is greater than or equal to the second threshold value, the input content is health content (discrimination) It may be determined as 260. In this case, the first threshold value is set to a value greater than the second threshold value, thereby increasing the accuracy of content determination. Conversely, when the maximum value of the plurality of first health content reliability 240 is less than the first threshold or the maximum value of the plurality of second health content reliability 250 is less than the second threshold, the input content is non-healthy content 265 ) Can be determined.

When the health content (discrimination) 260 is determined, the health content (discrimination) 260 is transmitted to the health keyword extraction unit 280, and the health keyword extraction unit 280 is A keyword (discrimination) 265 can be extracted.

The health keyword (discrimination) 265 is transmitted to the content reliability adjustment unit 295, and based on the health keyword (discrimination) 265, the first health content reliability (adjustment) 290 and the second health content reliability (adjustment) 285 can be determined. The content reliability adjustment unit 295 classifies only the health keyword (discrimination) 265 and determines the first health content reliability 240 and the second health content reliability 250 determined based on the existing primary health content 220. If adjustment is required, the adjusted reliability value may be transmitted to the health content determination unit 210.

Alternatively, the first health content reliability (adjustment) 290 and the second health content reliability (adjustment) 285 may include information related to adjustment for determining the reliability for each health keyword in the primary health content determination model 230. The first health content determination model 230 is based on the first health content reliability (adjustment) 290 and the second health content reliability (adjustment) 285 based on the first health content reliability based on the health keyword. The weight of the health keyword for determining 240 and the second health content reliability 250 may be adjusted.

3 is a conceptual diagram showing an operation of a content reliability adjustment unit according to an embodiment of the present invention.

In FIG. 3, a method for adjusting the reliability of the content reliability adjusting unit is disclosed.

Referring to FIG. 3, the content reliability adjuster may extract a separate reliability level for the health keyword (discrimination) of the health content (discrimination) from the health keyword extraction unit in order to adjust the existing reliability.

A health keyword (discrimination) 1 to a health keyword (discrimination) n may be extracted from each of the health content (discrimination) 1 to the health content (discrimination) n. A first health content reliability (discrimination) 310 and a second health content reliability (discrimination) 320 may be determined based on the extracted health keyword (discrimination) 1 to health keyword (discrimination) n.

The first health keyword reliability (discrimination) 310 and the second health keyword reliability (discrimination) 320 are the first health content reliability level 330 and the second health content level based on the primary health content determination model 350. Each of the reliability levels 340 may be compared, and the first health keyword reliability level (adjustment) 360 and the second health keyword reliability level (adjustment) 370 may be determined through the comparison.

The first health content reliability (adjustment) 360 and the second health content reliability (adjustment) 370 may be determined when the health content (discrimination) is greater than or equal to a threshold number.

Specifically, a separate secondary health content discrimination model 300 may be generated through separate learning on health content (discrimination) of a threshold number or more, and based on the secondary health content discrimination model 300, The first health content reliability (decision) 310 and the second health content reliability (decision) 320 may be determined.

The first health content reliability (discrimination) 310 and the second health content reliability (discrimination) 320 determined based on the secondary health content discrimination model 300 for the same health content, and the primary health content discrimination model 350 When the difference between the first health content reliability level 330 and the second health keyword reliability level 340 determined based on) is greater than or equal to the threshold value, the content reliability adjustment unit may determine that adjustment of the health keyword is necessary.

When adjustment of the health keyword is necessary, the reliability of the health content may be adjusted with the first health content reliability (adjustment) 360 and the second health content reliability (adjustment) 370. The adjusted first health content reliability (adjustment) 360 and the second health keyword reliability (adjustment) 370 are transmitted as feedback values to the health content determination unit and may be used to determine whether the input content is health content.

The first health content reliability (adjustment) 360 and the second health content reliability (adjustment) 370 are information for adjusting the reliability value for the corresponding health keyword and health keyword combination, and as described above, the health content determination unit 1 It is transmitted to the primary health content determination model and may be used to determine the reliability of the first health content and the reliability of the second health content later.

4 is a conceptual diagram illustrating a method for matching health content and medical information according to an embodiment of the present invention.

FIG. 4 discloses a method for determining medical information to be provided along with the health content to a user by matching health content (discrimination) and medical information when it is determined that the input input content is health content (discrimination).

Referring to FIG. 4, the machine learning unit may generate a learning model 460 for determining medical information by matching medical information 450 related to health content (discrimination) 400.

The machine learning unit may perform learning by receiving medical information for learning.

As described above, a combination of a plurality of health keywords and a plurality of health keywords may be determined in the health content (determination) 400, but the medical information 450 matching the health content (decision) 400 is a combination of a plurality of health keywords. It may be determined based on the health keyword combination 420 having the highest second health content reliability 410 among them. The accuracy of the determined medical information 450 may be improved by using the health keyword combination 420 having the highest reliability among the plurality of health keyword combinations. The health keyword combination having the highest second health content reliability 410 among the plurality of health keyword combinations may be expressed in terms of a health keyword combination (medical information) 420.

For example, the health keyword combination (medical information) 420 in specific health content may be {fever, cough, muscle pain}. In order to determine the health keyword combination (medical information) 420 and the medical information 450 having high relevance, a matching relationship between each medical information and the health keywords may be established.

When the medical information provided in connection with the health content is of a hospital type, information on a disease name treated in an internal medicine, otorhinolaryngology, surgery, or dermatology, and information on symptoms according to the disease name may be input in advance to perform learning. Health keyword combination (medical information) 420 is matched with any hospital type in consideration of the matching degree of the disease name and symptoms with the health keyword combination (medical information) 420 entered based on the learning model (medical information) 420 You can judge whether is high.

The degree of matching may be determined by setting weights for each disease name and symptom. According to an embodiment of the present invention, information on the names of diseases that mainly occur in each department can be learned, and the independent disease names and symptoms for each department will be adjusted so that the corresponding department and the health keyword (combination) have a higher degree of matching. I can. The degree of matching may be determined by adjusting the learning model to have a relatively low degree of matching for symptoms and disease names overlapping for each department.

For example, {fever, cough, muscle pain} can be matched with disease name (flu)-symptom (fever, cough, muscle pain), and information about internal medicine and otolaryngology that can treat flu can be provided as medical information. have.

If the matching degree between the health keyword combination (medical information) 420 and the medical information 450 is less than or equal to the threshold value, the matching degree with the medical information 450 is again based on the other subordinate health keyword combination (medical information) 440. It can be decided once. The lower-ranked health keyword combination (medical information) 440 may be a health keyword including n health keywords having a relatively high reliability among individual health keywords.

If the medical information 450 determined based on the health keyword combination (medical information) 420 and the medical information 450 determined based on the subordinate health keyword (medical information) 440 are the same, the medical information is It may be determined as medical information 450 matching the content (discrimination) 400.

If the medical information 450 determined based on the health keyword combination (medical information) 420 and the medical information 450 determined based on the subordinate health keyword combination (medical information) 440 are not the same, the health keyword combination ( The medical information 450 determined based on the medical information) 420 may be provided with priority, and the medical information 450 determined based on the combination of lower-ranked health keywords (medical information) 440 may be provided as a next priority. . For example, the medical information 450 determined based on the health keyword combination (medical information) 420 may be internal medicine and the medical information 450 determined based on the subordinate health keyword combination (medical information) 440 may be a dermatology have. In this case, if only one information window for health information is provided on the web page, information on internal medicine is provided first, and if two information window are provided, information on both internal medicine and dermatology is provided. Can provide.

Alternatively, according to an embodiment of the present invention, medical information matching each of the top n individual health keywords (medical information) having a relatively high reliability value as an input value may be determined. Medical information occupying a large number of matched medical information may be determined as medical information finally provided.

5 is a conceptual diagram showing a method of providing medical information according to an embodiment of the present invention.

5 discloses a method for a user to reserve a medical service based on provided medical information.

Referring to FIG. 5, medical information 520 may be provided on a health content webpage 500 provided on a user device provided by a medical information providing service server.

When the medical information is hospital information, the medical information 520 provided on the health content webpage 500 may include information on hospital treatment subjects, hospital location, and hospital reservation.

Medical information 520 having a high priority based on user information (eg, user location) may be provided on the health content webpage 500.

When the user wants to receive treatment for a related disease, the user may select the medical information 520 and make a hospital reservation.

For example, when the health content webpage 500 includes content related to an ophthalmic disease, the ophthalmic hospital information may be selected as the medical information 520, and information on the nearest ophthalmic hospital may be provided based on the user's location. I can.

The embodiments according to the present invention described above may be implemented in the form of program instructions that can 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 configured for the present invention or 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. medium), 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 can be changed to one or more software modules to perform the processing according to the present invention, and vice versa.

In the above, the present invention has been described by specific matters such as specific elements and limited embodiments and drawings, but this is provided only to help a more general understanding of the present invention, and the present invention is not limited to the above embodiments. Anyone with ordinary knowledge in the technical field to which the invention belongs can make various modifications and changes from these descriptions.

Accordingly, the spirit of the present invention is limited to the above-described embodiments and should not be defined, and all ranges equivalent to or equivalently changed from the claims to be described later as well as the claims to be described later are the scope of the spirit of the present invention. It will be said to belong to.

Claims (6)

The method of providing medical information related to health content based on artificial intelligence,
Determining, by the medical information providing service server, health content;
Determining, by the medical information providing service server, medical information associated with the health content; And
Including the step of the medical information providing service server transmitting the medical information to the medical information receiving server,
The medical information receiving server provides the health content and the medical information related to the health content together,
The health content is determined based on a plurality of first health content reliability for each of a plurality of health keywords extracted based on a primary health content determination model and a plurality of second health content reliability levels for each of the plurality of health keyword combinations Become,
The health content is a content in which a maximum value of the reliability of the plurality of first health content is equal to or greater than a first threshold value, and a maximum value of the reliability of the plurality of second health content is equal to or greater than a second threshold value,
The reliability of the first health content and the reliability of the second health content are determined based on a weight set for each of a plurality of health keywords,
The weight is set to a relatively high value for a specialized medical term based on learning a separate medical term in the primary health content discrimination model, and a relatively high value for a medical term with a relatively low frequency of use among the medical terms. A method, characterized in that it is set to a value. The method of claim 1,
Content reliability adjustment unit for the first health content reliability determined based on the primary health content determination model, the second health content reliability, and health content equal to or greater than a threshold number determined as health content by the primary health content determination model. If the difference between the reliability of the first health content (discrimination) and the reliability of the second health content (discrimination) determined based on each independently generated secondary health content discrimination model based on separate learning is a threshold, the first health And generating the content reliability and the second health content reliability by adjusting each of the first health content reliability (adjustment) and the second health content reliability (adjustment). The method of claim 2,
In the medical information, a health keyword combination (medical information), which is a health keyword combination having a maximum value among the plurality of second health contents reliability, and information on a disease name and symptoms according to the disease name are input in advance to a learning model ( Is determined based on the matching degree of disease name and symptom),
The method according to claim 1, wherein the matching degree is adjusted so that the corresponding medical department and health keyword (combination) have a higher degree of matching as the independent disease name and independent symptom for each department. A medical information providing service server that provides medical information related to health content based on artificial intelligence,
A communication unit implemented to receive input content; And
A processor operatively connected to the communication unit,
The processor determines health content among the input content,
Determine medical information associated with the health content,
It is implemented to transmit the medical information to the medical information receiving server,
The medical information receiving server provides the health content and the medical information related to the health content together,
The health content is determined based on a plurality of first health content reliability for each of a plurality of health keywords extracted based on a primary health content determination model and a plurality of second health content reliability levels for each of the plurality of health keyword combinations Become,
The health content is a content in which a maximum value of the reliability of the plurality of first health content is equal to or greater than a first threshold value, and a maximum value of the reliability of the plurality of second health content is equal to or greater than a second threshold value,
The reliability of the first health content and the reliability of the second health content are determined based on weights set for each of a plurality of health keywords,
The weight is set to a relatively high value for a specialized medical term based on learning a separate medical term in the primary health content discrimination model, and a relatively high value for a medical term with a relatively low frequency of use among the medical terms. Medical information providing service server, characterized in that set to a value. The method of claim 4,
Content reliability adjustment unit for the first health content reliability determined based on the primary health content determination model, the second health content reliability, and health content equal to or greater than a threshold number determined as health content by the primary health content determination model. If the difference between the reliability of the first health content (discrimination) and the reliability of the second health content (discrimination) determined based on each independently generated secondary health content discrimination model based on separate learning is a threshold, the first health A medical information providing service server, characterized in that the content reliability and the second health content reliability are adjusted to each of the first health content reliability (adjustment) and the second health content reliability (adjustment). The method of claim 5,
In the medical information, a health keyword combination (medical information), which is a health keyword combination having a maximum value among the plurality of second health contents reliability, and information on a disease name and symptoms according to the disease name are input in advance to a learning model ( Is determined based on the matching degree of disease name and symptom),
The medical information providing service server, characterized in that the matching degree is adjusted to have a higher matching degree between the corresponding medical department and the health keyword (combination) as the independent disease name and independent symptom for each department.

Images