KR102451066B1 - System for providing critical value report using text mining technique and method thereof - Google Patents

Abstract

The present invention relates to a system and method for providing important inspection results using a text mining technique. According to the present invention, in a method for providing a critical test result using a system for providing a critical test result to which a text mining technique is applied, storing keywords corresponding to a Critical Value Report (CVR) in a database; The steps of receiving a reading for a subject to be examined by a medical examiner, extracting each keyword present in the reading using text mining, the extracted plurality of keywords and pre-stored important test results (CVR) ), and if the extracted keyword matches the keyword corresponding to the critical test result (CVR), providing the keyword as the reading text to the reader.
As described above, according to the present invention, it is possible to read an image of a subject to be examined in real time while not missing the presence or absence of an important examination result among the reading result report of a radiology examination. In addition, when a signal approved as an important test result is received from a radiologist, it is quickly provided to the attending physician, clinician, and specialist in the department closely related to the important test result, so that emergency measures can be taken quickly for the test subject.

Description

A system for providing critical inspection results using text mining techniques and a method therefor

The present invention relates to a system for providing examination results using a text mining technique and a method therefor. The present invention uses a text mining technique that extracts keywords from a reading text, checks the extracted keywords and important test results, and provides them to related medical staff in real time. It relates to a system for providing inspection results and a method therefor.

In clinical situations, the report written by the radiologist is an important medium through which information can be communicated and opinions can be exchanged with the clinician who has requested the examination.

In this process, a test result that has a very clinical significance or requires immediate action is called a Critical Value Report (CVR), which is a very important factor in the diagnosis of a patient and should be reported immediately. It consists of urgent medical findings and a quick CVR report is an essential element for the rapid establishment of a treatment plan and the execution of a prompt and appropriate treatment process.

Currently, the report of CVR in the field of radiology is based on the computerized system that notifies the prescribing physician/physician by PACS/EMR, etc., on the basis that the interpreter has recognized the contents, and if the system is not equipped This is done in a passive way to report directly over the wire.

However, in a situation in which radiology examinations are exploding due to the development of medical technology and the increase in medical access, the radiologist who performed the reading lacks understanding of CVR or CVR after writing the report even though he has knowledge about the situation. If the contents are ignored or overlooked, the opportunity to start reporting the CVR may be missed even if an appropriate computer system is in place in advance.

In addition, important findings in the clinical situation may be overlooked, resulting in fatal results for the patient, and in some cases may even lead to legal problems.

The technology that is the background of the present invention is disclosed in Korean Patent Registration No. 10-1963605 (2019.03.29. Announcement).

The technical problem to be achieved by the present invention is to provide a system and method for providing examination results using a text mining technique that extracts keywords from a reading text, checks the extracted keywords and important test results, and provides them to relevant medical staff in real time. .

According to an embodiment of the present invention for achieving this technical problem, in a method for providing a critical test result using a system for providing a critical test result to which a text mining technique is applied, keywords corresponding to a Critical Value Report (CVR) are stored in a database. storing the medical-related image, receiving an input of a reading for a subject to be examined by a medical examiner with respect to the medical-related image, extracting each keyword present in the reading using text mining, the extracted plurality of keywords and Comparing with a keyword corresponding to a pre-stored critical test result (CVR), if the extracted keyword matches a keyword corresponding to the critical test result (CVR), the keyword is sent to the reader like the reading text step of providing.

Upon receiving an authentication signal for the corresponding keyword provided by the reading doctor, providing the keyword and the reading to at least one user terminal among a clinician, a primary care physician, and a specialist in a department related to the important test result of the examination subject; may include more.

counting the number of times that the keyword has been input as a non-authentication signal when a non-authentication signal is received from the reading intent for the corresponding keyword provided with the reading intent, and when the counted number reaches a reference number of times, the corresponding keyword It may further include the step of deleting from the important test result keyword.

In the step of extracting each keyword present in the read text, the keyword may be extracted by performing morphological analysis or semantic analysis on the read text.

calculating a degree of similarity between the extracted keyword and the keyword corresponding to the important test result, and if the calculated similarity is greater than or equal to a reference value, determining the extracted keyword as a typo and changing it to a keyword corresponding to the important test result The method may further include the step of providing the reading to the reader like the reading.

The calculating of the similarity may be performed using the following equation.

In the storing in the database, keywords corresponding to the CVR may be matched and stored in the form of codes.

The medical-related image may include any one of an ultrasound image, a computed tomography (CT) image, a magnetic resonance image (MRI), and an X-ray image. .

According to another embodiment of the present invention, in a system for providing a critical test result using a text mining technique, a database for storing keywords corresponding to a Critical Value Report (CVR) and a medical-related image are read by a doctor An input unit that receives the written reading for the subject to be examined, an extraction unit that extracts each keyword present in the reading using text mining, the plurality of extracted keywords and the pre-stored important test results (CVR) and a comparison unit that compares the keywords, and a control unit that, when the extracted keyword matches the keyword corresponding to the critical test result (CVR), provides the keyword as the reading text to the reader.

As described above, according to the present invention, it is possible to read an image of a subject to be examined in real time while not missing the presence or absence of an important examination result among the reading result report of a radiology examination.

In addition, when a signal approved as an important test result is received from a radiologist, it is quickly provided to the attending physician, clinician, and specialist in the department closely related to the important test result, so that emergency measures can be taken quickly for the test subject.

1 is a configuration diagram for explaining an important inspection result providing system according to an embodiment of the present invention.
FIG. 2 is a configuration diagram for explaining the configuration of the control device of FIG. 1 .
3 is a flowchart for explaining a method for providing an important inspection result using a system for providing an important inspection result according to an embodiment of the present invention.
4 is a view showing a critical test result (CVR) stored in a database according to an embodiment of the present invention.
FIG. 5 is an exemplary diagram of a read statement for explaining step S320 of FIG. 3 .
6 is an exemplary view showing a screen for providing CVR information to a reader's terminal according to an embodiment of the present invention.
7 is an exemplary diagram illustrating a screen for providing CVR information to a terminal of a attending physician or a clinician 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 carry out the present invention. However, the present invention may be embodied in several different forms and is not limited to the embodiments described herein. And in order to clearly explain the present invention in the drawings, parts irrelevant to the description are omitted, and similar reference numerals are attached to similar parts throughout the specification.

Throughout the specification, when a part "includes" a certain component, it means that other components may be further included, rather than excluding other components, unless otherwise stated.

Then, with reference to the accompanying drawings, embodiments of the present invention will be described in detail so that those of ordinary skill in the art to which the present invention pertains can easily implement them.

Hereinafter, a system for providing important inspection results according to an embodiment of the present invention will be described with reference to FIGS. 1 and 2 .

1 is a configuration diagram for explaining an important inspection result providing system according to an embodiment of the present invention.

As shown in FIG. 1 , the system for providing important test results includes a database 100 , a control device 200 , a plurality of reading terminals 300 , a doctor terminal 400 , and a clinician terminal 500 .

First, the database 100 may be created by a separate management server, and stores keywords corresponding to the Critical Value Report (CVR).

Here, the Critical Value Report (CVR) is a test result that has a very clinical significance or requires immediate action, and consists of important and urgent medical findings that must be reported immediately.

Next, the control device 200 receives the written reading from the reader, extracts keywords from the inputted reading, determines whether the keyword corresponds to the important test result, and provides it to the reader.

In addition, the plurality of reading terminals 300 , the doctor's terminal 400 and the clinician's terminal 500 are substantially the same type of user terminals, and for convenience of explanation, the reading terminal 300 and the doctor's The terminal 400 is called a clinician terminal 500 .

In addition, the reading terminal 300, the doctor's terminal 400, and the clinician's terminal 500 include a portable user terminal, and are wired or It can be implemented as a device capable of wirelessly accessing a network and exchanging information.

FIG. 2 is a configuration diagram for explaining the configuration of the control device of FIG. 1 .

As shown in FIG. 2 , the control device 200 includes an input unit 210 , an extraction unit 220 , a comparison unit 230 , and a control unit 240 .

First, the input unit 210 receives a reading text for a subject to be examined written by a medical doctor with respect to a medical-related image.

Here, the readout includes essential content for the medical-related image by the doctor.

In addition, the medical-related image includes any one of an ultrasound (Ultrasound) image, a computed tomography (CT) image, a magnetic resonance image (MRI), and an X-ray image.

Next, the extraction unit 220 extracts each keyword present in the read text using a text mining method.

At this time, in order to extract the keywords present in the read text, the extraction unit 220 extracts the keywords by performing morphological analysis or semantic analysis.

Here, the morpheme analysis is a method to grasp the structure of various linguistic properties such as morphemes, roots, prefixes/suffixes, and parts of speech (POS, part-of-speech), and semantic analysis interprets the meaning of the document, A search result desired by the user is provided by analyzing the meaning and context of the vocabulary described in . In the present invention, morphological analysis or semantic analysis is used to extract keywords present in the read text.

Next, the comparison unit 230 compares the plurality of extracted keywords with the plurality of keywords corresponding to the previously stored CVR.

Then, when the extracted keyword matches the keyword corresponding to the critical test result (CVR), the control unit 240 provides the keyword together with the read text to the reading terminal 300 .

Hereinafter, an important inspection result method using the important inspection result providing system will be described with reference to FIGS. 3 to 7 .

3 is a flowchart for explaining a method for providing an important inspection result using a system for providing an important inspection result according to an embodiment of the present invention.

First, the database 100 according to an embodiment of the present invention stores keywords corresponding to a Critical Value Report (CVR) (S310).

4 is a view showing a critical test result (CVR) stored in a database according to an embodiment of the present invention.

That is, as shown in FIG. 4, it includes a section grouped between observations and similar observations corresponding to the CVR results and a classification code corresponding to each CVR.

Next, the control device 200 receives a reading text for a subject to be examined written by a medical examiner with respect to a medical-related image (S320).

FIG. 5 is an exemplary diagram of a read statement for explaining step S320 of FIG. 3 .

That is, the reader writes the result of reading the medical-related image in the form of a reading as shown in FIG. 5 , and inputs the reading through the reader terminal 300 .

Then, the control device 200 receives the reading text for the subject to be examined written by the reading doctor through the reading text terminal 300 .

Although the reading text is shown in the form of a sentence in FIG. 5, it may be written in various sentence or syntax forms according to the reader and the program used.

Next, the control device 200 extracts each of the keywords present in the read text input from step S320 through the text mining technique ( S330 ).

For example, for the readout shown in FIG. 5 , the control device 200 may display 'thromboembolism', 'granuloma', 'pulmonary', 'lobar', 'branch', 'calcified', 'right', 'upper', You can extract 'in' and 'left' as keywords.

Here, 'thromboembolism' and 'granuloma' are keywords related to disease, 'pulmonary', 'lobar', 'branch', and 'calcified' are keywords related to anatomical terms, and 'right', 'upper', 'in' , 'left' is a keyword for other adjectives or words.

Then, the control device 120 compares the keyword extracted in step S330 with the CVR pre-stored in the database to determine whether the keyword matches ( S340 ).

That is, the control device 200 compares the extracted keyword with a plurality of keywords corresponding to the CVR shown in FIG. 4 .

In this case, the control window 120 calculates a degree of similarity between the extracted keyword and the keyword corresponding to the important test result in order to determine whether the extracted keyword includes a typo.

Here, the control window 120 calculates the degree of similarity between the extracted keyword and the keyword corresponding to the CVR using Equation 1 below.

At this time, if the degree of similarity calculated through Equation 1 is equal to or greater than the reference value, the control device 200 determines the extracted keyword as a typo and changes it to a keyword corresponding to an important test result.

For example, if the extracted keyword is 'thromboemboliam' and the keyword corresponding to the stored critical test result (CVR) is 'thromboembolism', since one letter is different, the similarity is 0.93 (14/15) by Equation 1 is calculated

If it is assumed that the reference value is 0.8, in the case of the extracted keyword 'thromboemboliam', the similarity is calculated as 0.93, and since the calculated similarity is greater than the reference value, the controller 200 sets 'thromboemboliam' as a typo for thromboembolism. judge

Accordingly, the control device 200 converts the extracted keyword 'thromboemboliam' into 'thromboembolism', a keyword corresponding to the CVR.

On the other hand, when the extracted keyword is 'thrombainboliam', the similarity calculated through Equation 1 is calculated as 0.73 (11/15), and it is determined that the extracted keyword is not a keyword corresponding to the critical test result (CVR) and is terminated. do.

Next, when the extracted keyword is included in the critical test result (CVR), the control device 200 provides the keyword corresponding to the critical test result (CVR) to the reading terminal 300 together with the reading written by the reader. do (S350).

6 is an exemplary view showing a screen for providing CVR information to a reader's terminal according to an embodiment of the present invention.

That is, as shown in FIG. 6 , the control device 200 provides the keyword corresponding to the critical test result (CVR) to the reading terminal 300 in a pop-up format.

Next, the control device 200 receives an authentication signal for determining whether the corresponding keyword is a critical test result (CVR) from the reading terminal 300 (S360).

That is, the reader selects any one of “YES” and “NO” shown in FIG. 6 through the reading terminal 300 and transmits an authentication signal or a non-authentication signal to the control device 200 .

Here, when the corresponding keyword corresponds to the critical test result (CVR), but the doctor determines that the relevance to the symptoms of the current examination subject is low, the doctor transmits the non-authentication signal to the control device 200 through the reader's terminal send.

At this time, if the reader transmits the non-authentication signal using the reading terminal 300, the control device 200 counts the number of times the corresponding keyword is input as the non-authentication signal (S370).

That is, the control device 200 counts the number of times the intention of reading input the non-authentication signal with respect to the corresponding keyword extracted as the critical test result (CVR).

Next, when the counted number reaches the reference number, the control device 200 deletes the corresponding keyword from the CVR keyword (S380).

That is, the control device 200 deletes the corresponding keyword from the CVR keyword when the number of times the number of times the non-authentication signal is input by the reading intention reaches the reference number.

For example, if the number of counts of non-authentication signals of 'thromboembolism', a keyword corresponding to the critical test result (CVR), reaches the reference 10 times, the keyword corresponding to the critical test result (CVR) in the database 100 is Delete and update 'thromboembolism'.

On the other hand, when the reading doctor transmits an authentication signal using the reading terminal 300, the control device 200 communicates with the patient's primary doctor terminal 400, the clinician's terminal 500, and the important test result (CVR). A corresponding keyword and a reading are provided to a terminal (not shown) of a specialist in a medical department having a close relationship ( S390 ).

For example, it is assumed that the patient's clinician belongs to the department of respiratory medicine, and the patient's chest CT is requested to be read by the radiologist affiliated with the radiology department. If the 'thromboembolism' keyword corresponding to the critical test result (CVR) exists in the reading written by the reader, the control device 200 transmits the keyword and the reading to the clinician's terminal as well as the 'thromboembolism' keyword and The relevant keywords and readings are also sent to gastroenterologists who are closely related.

7 is an exemplary diagram illustrating a screen for providing CVR information to a terminal of a attending physician or a clinician according to an embodiment of the present invention.

That is, as shown in FIG. 7 , the critical test result (CVR) in the form of an SMS is sent to the doctor's terminal 400, the clinician's terminal 500, and the medical department's terminal (not shown) related to the critical test result (CVR). ) is provided.

Although it is shown in FIG. 7 that the critical inspection result (CVR) is provided in the form of an SMS, it may be provided in a form such as a pop-up form or a mail form depending on the system 100 for providing the important inspection result.

As described above, according to an embodiment of the present invention, it is possible to read an image of a subject to be examined in real time while not missing the presence or absence of an important examination result among the reading result report of a radiology examination.

In addition, when a signal approved as an important test result is received from a radiologist, it is quickly provided to the attending physician, clinician, and specialist in the department closely related to the important test result, so that emergency measures can be taken quickly for the test subject.

Although the present invention has been described with reference to the embodiment shown in the drawings, this is merely exemplary, and those skilled in the art will understand that various modifications and equivalent other embodiments are possible therefrom. Accordingly, the true technical protection scope of the present invention should be defined by the technical spirit of the appended claims.

100: database, 200: control unit,
210: input unit, 220: extraction unit,
230: comparison unit, 240: control unit,
300: a reading terminal, 400: a doctor's terminal;
500: clinician terminal

Claims (16)

In a method for providing important inspection results using a system for providing important inspection results to which text mining techniques are applied,
Storing keywords corresponding to the Critical Value Report (CVR) in the database;
A step of receiving an examination text for a subject for examination written by a medical examiner with respect to a medical-related image;
extracting each keyword present in the readout using text mining;
Comparing the extracted keywords with keywords corresponding to the previously stored CVR;
if the extracted keyword matches the keyword corresponding to the critical test result (CVR), providing the keyword as the reading text to the reader;
When an authentication signal is input for the corresponding keyword provided by the reading doctor, providing the keyword and the reading to at least one user terminal among a clinician, a primary care physician, and a specialist in a department related to the important test result of the examination subject;
counting the number of times the keyword is input as a non-authentication signal when a non-authentication signal is received from the reading intent for the corresponding keyword provided with the reading intent; and
and deleting the corresponding keyword from the important test result keyword when the counted number reaches a reference number,
The step of extracting each keyword present in the reading comprises:
A method for providing an important test result for extracting keywords by performing morphological or semantic analysis on the read text. delete delete delete According to claim 1,
calculating a degree of similarity between the extracted keyword and the keyword corresponding to the important test result; and
and if the calculated similarity is greater than or equal to a reference value, determining the extracted keyword as a typo, changing the keyword to a keyword corresponding to the important test result, and providing the extracted keyword to the reading doctor as the reading text. 6. The method of claim 5,
The step of calculating the similarity is:
A method of providing important test results calculated using the following equation.

According to claim 1,
The step of storing in the database,
A method of providing a critical test result by matching the keywords corresponding to the critical test result (CVR) in a code form and storing the matching. According to claim 1,
The medical image is
An ultrasound (Ultrasound) image, a computed tomography (CT) image, a magnetic resonance image (MRI), and a method for providing important examination results including any one of an X-ray image. In the system for providing important inspection results using text mining technique,
A database that stores keywords corresponding to the Critical Value Report (CVR);
An input unit for receiving an examination text for a subject to be examined written by a medical examiner with respect to a medical-related image;
an extracting unit for extracting each keyword present in the reading using text mining;
a comparison unit for comparing the extracted keywords with keywords corresponding to the previously stored important test results (CVR); and
When the extracted keyword matches the keyword corresponding to the critical test result (CVR), the control unit provides the keyword to the reader as the reading text,
The control unit is
When an authentication signal is received for the corresponding keyword provided by the reading doctor, the keyword and the reading are provided to at least one user terminal among a clinician, a primary care physician, and a specialist in a department related to the important test result of the examination subject,
When a non-authentication signal is received from the reading intention for the corresponding keyword provided with the reading intent, the number of times the corresponding keyword is input as a non-authorization signal is counted. Delete it from the resulting keywords,
The extraction unit,
An important test result providing system for extracting keywords by performing morphological or semantic analysis on the read text. delete delete delete 10. The method of claim 9,
The control unit is
The degree of similarity between the extracted keyword and the keyword corresponding to the important test result is calculated, and if the calculated similarity is greater than or equal to a reference value, the extracted keyword is determined as a typo and changed to a keyword corresponding to the important test result, and the A system for providing important test results to the doctor, such as a report. 14. The method of claim 13,
The control unit is
An important inspection result providing system that calculates using the following equation.

10. The method of claim 9,
The database is
A system for providing important inspection results that matches and stores keywords corresponding to the important inspection results (CVR) in a code form. 10. The method of claim 9,
The medical image is
A system for providing important test results including any one of an Ultrasound image, a Computed Tomography (CT) image, a Magnetic Resonance Image (MRI), and an X-ray image.

Images