KR101692930B1 - Medical Record Translation System and Medical Record Translation Method - Google Patents

Abstract

The present invention relates to a system for translating a medical record written with a medical term and a translation method thereof. The present invention has the following effects. The present invention can improve the accuracy of translation by performing a searching operation according to a priority. Also, a user can optionally correct a translation result with a translation unit and construct a database by storing the corrected translation, thereby improving a translation rate according as the experience of the user is accumulated. The present invention includes a medical record scan module which converts a medical record scan material into a medical record text material, a medical record translation module which translates the medical record text material from English to Korean, and the database which stores data to be translated.

Description

Medical Record Translation System and Medical Record Translation Method {omitted}

The present invention relates to a system for translating medical recording papers written in medical terminology in real time and a translation method thereof.

The purpose of this study is to investigate and evaluate the amount of loss and the assessment of insurance premiums by objective and fair and qualified professionals who are important business affects not only the insurance companies but also the policyholders' interests. The rationality, In order to improve the soundness and fairness of the business, a system of damages assessment was introduced.

The role of the damage assessor is to provide the insurer and the insured side with the insurance claim payment liability and the agreed insurance standard amount through the accident investigation, the judging of the book and the process, the fair damages, It is to be resolved within the insurance system.

The damage assessment process is shown in FIG. 7, where the insurer receives a claim from the customer and begins by reviewing the documents for the receipt and requesting an investigation to the loss assessment company. Damage Assessment The Company receives the request for investigation and distributes the request for each person in charge according to the nature and contents of the investigation request. The designated investigator will analyze the investigation, review the data, and prepare a report, which includes medical records such as medical certificates and certificates issued by the hospital. However, most of the medical records are in English medical terminology, and those who do not have expertise in the field of medicine as a doctor are very time consuming to interpret it. Especially medical terminology is not listed in the general English dictionary. Therefore, it is necessary to translate dictionary of medical field one by one, and it is troublesome work such as searching for the word which is not an abbreviation or official term, and grasping it. Therefore, it usually takes 5 to 10 days for researchers to analyze medical records from dozens to hundreds.

Registration No. 10-1313158 (September 30, 2013)

The problems to be solved in the present invention are as follows.

We want to improve the accuracy of translation by performing searches according to priority. In this paper, we propose an intelligent translation system that enables users to make arbitrary revisions for each translation unit in a translation result, stores it in a database, and increases the translation rate as the user experience accumulates.

In order to solve the above problems,

A medical record scanning module (100) for converting the medical record scan book (10) into a medical record text book (20); A medical recording paper translation module 200 for translating the English text of the medical recording paper text 20 into Korean; A database 300 in which data to be translated is stored; And a translation creating module (400) for creating a medical recording translation (30) by reflecting the translated result by the medical recording translation module (200) in the medical record text book (20) And the medical record translation module 200 includes a sentence code or a unit sentence separated by a line break in the medical record paper text block 20. The medical record translation module 200 may include a medical phrase database 350 in which English medical terms and corresponding Hangul phrases are stored, A translation unit 220 for performing translation for each translation unit extracted according to a predetermined order, and a translation output unit 230 for outputting translation results for each translation unit, The extraction unit 210 includes an abbreviation extraction unit 214 for extracting an abbreviation in a unit sentence, a phrase extraction unit 211 for extracting the longest translatable phrase in a unit sentence, an abbreviation extraction unit 214, (Not shown) A user word extracting unit 212 for extracting a user word among the texts extracted from the unit word extracting unit 214, a phrase extracting unit 211, and a user word extracting unit 212, The vocabulary extracting unit 213 extracts a vocabulary among the texts not extracted by the learning word extracting unit 213, the abbreviation extracting unit 214 and the phrase extracting unit 211, the user word extracting unit 212 and the learning word extracting unit 213, (Punctuation marks, etc.) or non-words (e.g., punctuation marks) that are not extracted by the extracting unit 215, the abbreviation extracting unit 214 and the phrase extracting unit 211, the user word extracting unit 212 and the learning word extracting unit 213, And a personal information extraction unit 219 for extracting personal information from the text of the translation target file, wherein the user word is an English word directly registered by the user, the learning word is a translated word Wherein the meaning is an English word modified by the user, If the whole unit sentence is stored in the phrase database 350, it is extracted as a phrase. Otherwise, the phrase stored in the phrase database 350 is stored in the phrase database 350 while decreasing the length of the phrase in a manner of excluding the last word And extracts the phrases as phrases when they are stored in the phrase DB 350 while sequentially reducing the length of the phrases in the remaining words when one phrase is extracted, An abbreviation translating unit 224 for translating the abbreviations extracted by the user extraction unit 214, a phrase translation unit 221 for translating the phrases extracted by the phrase extraction unit 211, a user word extracted from the user word extraction unit 212, A learning word translation unit 223 for translating the learning word extracted by the learning word extraction unit 213, a vocabulary translation unit for translating the vocabulary extracted by the vocabulary extraction unit 215 225), the word extracted by the word extracting unit 216 is translated And the translated sentence output unit 230 includes a sentence output unit 230 for outputting a sentence sentence 231a extracted as a unit sentence by the extracting unit 210 and a translated sentence 231b corresponding thereto, A detailed unit output unit 232 for outputting a source word 232a dividing the sentence sentence 231a into a detailed translation unit and a translation word 232b corresponding to the source word 232b in the same segment, And a translation information display unit 233 for displaying translation information including the number of non-target words, the number of unregistered words, and the translation ratio, wherein the detailed translation unit is one of a phrase, a user word, a learning word, an abbreviation and a vocabulary, A translation is performed, and the translation word (232b) is abbreviated for each detailed translation unit, an automatically translated word including a vocabulary and a user word, a manually translated word including a learning word, an untranslated word, Display different colors for different words The translating sentence 231b is reflected in the translation sentence 231b and the translated sentence creation module 400 substitutes the sentence sentence 231a at each corresponding position of the medical recording paper text book 20, And a medical recorder translation system which records the medical recorder translation system.

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

The present invention has the following effects.

It is possible to improve the accuracy of the translation by searching according to the priority. In addition, the translation result can be arbitrarily modified by the user for each translation unit, stored in a database, and the translation rate is increased as the user experience is accumulated.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a diagram showing a screen on which translation is performed according to the present invention; FIG.
2 is an exemplary illustration of a medical recording sheet to be translated;
Fig. 3 is an exemplary view of a result of translating the medical recording sheet of Fig. 2; Fig.
4 is a diagram of a user word list;
5 is a block diagram of an overall configuration of the present invention;
6 is a block diagram of the detailed configuration of the medical record translation module;
7 is a diagram of a damage assessment process;
FIG. 8 is a view showing the medical recording sheets of FIGS. 2 and 3 on one screen. FIG.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings. However, the scope of the present invention should be understood from the description of the claims. Further, the description of known technology which obscures the gist of the present invention is omitted.

The medical recording paper translation system of the present invention includes a medical recording paper scanning module 100, a medical recording paper translation module 200, a database 300 and a translation writing module 400 as shown in FIG.

The medical recording sheet scanning module 100 provides a function of converting the medical recording sheet scanned book 10 into a medical recording sheet text book 20 and includes a scan file receiving unit 110, a file format converting unit 120, a text extracting unit 130, and a text file generation unit 140. The medical record paper text 20 is a file in which the medical record scanned book 10 is converted into a file converted by the medical record scanning module 100, .

The scan file receiving unit 110 is a part for receiving a scan file 10, that is, a file generated by scanning a medical record. At this time, medical records were issued by the hospital, and most of them are written in medical terminology in English. When a medical recording sheet is scanned, a medical record sheet scanned image 10, which is an image file in which an entire image is processed in an image format, is generated without recognizing the text written on the medical recording sheet directly. When the medical record book 10 is uploaded to the server to be translated, the scan file receiver 110 receives the scanned book 10. That is, when the user uploads a scanned medical record (10, e.g., a jpg file) generated by scanning the hardcopy medical recording sheet to the server, the scan file receiving unit 110 receives the scan.

The file format conversion unit 120 converts the scanned medical record 10 received from the scan file receiving unit 110 into a file format that can be extracted as text. The format of the text extractable file will depend on the program (tool) from which the text is to be extracted, but a pdf or xml file format that would normally perform OCR functions would be desirable. That is, the jpg file received from the scan file receiving unit 110 is converted from the file format conversion unit 120 into an xml file.

The text extraction unit 130 extracts an image from a file format-converted by the file format conversion unit 120 and extracts it as text. Since the conventional OCR function is used, detailed description is omitted.

The text file generating unit 140 is a unit for storing the text extracted by the text extracting unit 130 as a text file 20 of a medical record text to be translated. The medical record text text 20 may be a txt file containing only text information but may correspond to a location where the text extracted from the medical record scanned book 10 is extracted so as to be easily collated with the scanned medical record book 10 It is preferable to store the text position information and the portion not extracted as text in an image form. Further, the present invention is characterized by a web-based real-time translation, and an html file that can be easily retrieved from the web is more preferable. In other words, the text extraction unit 130 extracts text from the pdf file converted by the file format conversion unit 120, and then stores the extracted text information in the html file including the text information extracted by the text file generation unit 140 .

FIG. 2 shows an example of a medical recording sheet to be translated. This is a medical record sheet text block 20 converted by the medical recording sheet scanning module 100 into a scanned medical record sheet 10, It would also be a document of the same type.

The medical record translation module 200 provides a function of translating the English text of the medical record text text 20 into Korean and includes an extraction unit 210, a translation unit 220, and a translation output unit 230 .

The extraction unit 210 extracts a translation unit from the medical record text text 20 as a translation target file and includes a phrase extraction unit 211, a user word extraction unit 212, a learning word extraction unit 213, An extracting unit 214, a lexical extracting unit 215, a word extracting unit 216, an unregistered extracting unit 217, a non-object extracting unit 218 and a personal information extracting unit 219.

The user word means an English word registered directly by the user. If the English word to be translated is a word not registered in the database 300, the corresponding Korean word can not be called up in the database 300, so that the user directly inputs or selects the appropriate Korean word by searching the web . The input Hangul word and the English word to be translated correspond to the user word, and are stored in the user word DB 320.

The learning word means the English word modified by the user means the translated word. If the English word to be translated is a word registered in the database 300, the corresponding Korean word is called up in the database 300 and output by the translated text output unit 230. In this case, when the user desires to input a Korean word which is different from the output of the Korean word meaning, the modified Korean word and the English word to be translated correspond to the learning word. The learning word is stored in the learning word DB 330. [

A translation unit is a sentence unit that extracts the text of a file to be translated in sentence units, and it is a sentence code such as a period ('.') Or a unit sentence separated by a line break. In particular, due to the nature of the medical recording paper, since a unit sentence is often a list of words for a specific item rather than a complete sentence of a description form, the unit sentence of the present invention generally means a grammatically prepared sentence composed of a subject word and a descriptive word no.

The abbreviation extracting unit 214 extracts abbreviations from the unit sentence. Among the text of the unit sentence, the abbreviation dictionary DB 340 is searched for an abbreviation stored therein and extracted.

The phrase extraction unit 211 extracts the longest translatable phrase in the unit sentence. Among the text of the unit sentence, the phrase DB 350 is searched for a phrase stored therein and extracted. However, if the unit sentences are stored in the phrase DB 350, the sequence is first extracted. If not, the phrases stored in the phrase DB 350 are sequentially stored in a manner that the lengths of the phrases are sequentially reduced, . When one phrase is extracted, the length of the phrase is sequentially reduced in the remaining words, and it is determined whether the phrase is stored in the phrase database 350. [

For example, when a sentence is composed of "A, B, C, D", the sentence itself is not stored in the phrase DB 350, but "A, B, C, "And" C, D "are stored in the phrase database 350, respectively.

First, "A, B, C, D" except for the last word "E" is searched in the phrase DB 350. However, since it is not stored, it also searches for "A, B, C" whose length is reduced except for the last word "D". Quot; A " and " B "except for" C " Then, the remaining words "C, D, E" are searched in the phrase database 350, and if there is no end word "C, D" excluding the last word "C, D"

The user word extracting unit 212 extracts user words among texts not extracted by the abbreviation extracting unit 214 and the phrase extracting unit 211 in a unit sentence. In the unit sentence, the abbreviation extracting unit 214 and the phrase extracting unit 211 search for a user word stored in the user word DB 320 among texts not extracted, and extract the extracted user word.

The learning word extracting unit 213 extracts a learning word among texts not extracted by the abbreviation extracting unit 214, the phrase extracting unit 211, and the user word extracting unit 212 in a unit sentence. A unit sentence is searched for a learning word stored in the learning word DB 330 among the texts not extracted by the abbreviation extraction unit 214, the phrase extraction unit 211, and the user word extraction unit 212, and extracted.

The vocabulary extraction unit 215 is a unit for extracting vocabulary among texts not extracted by the abbreviation extraction unit 214, the phrase extraction unit 211, the user word extraction unit 212, and the learning word extraction unit 213. It is checked whether there is a vocabulary stored in the English dictionary DB 310 among the texts not extracted by the abbreviation extracting unit 214, the phrase extracting unit 211, the user word extracting unit 212 and the learning word extracting unit 213 in the unit sentence And extracts it.

The unregistered extracting unit 217 extracts an unregistered word among texts not extracted by the abbreviation extracting unit 214, the phrase extracting unit 211, the user word extracting unit 212, and the learning word extracting unit 213, The non-target extraction unit 218 extracts non-target words from the non-extracted text in the abbreviation extraction unit 214, the phrase extraction unit 211, the user word extraction unit 212, and the learning word extraction unit 213 Etc.) and extracts Hangul.
The personal information extracting unit 219 extracts the personal information from the text of the translation target file. Personal information may include information such as a resident registration number. In the case of a resident registration number, recognition methods such as recognizing the case where the digits of 6 or more digits are arranged with a hyphen ('-') between them can be configured in various ways.

delete

The translation unit 220 is a unit for performing translation for each translation unit extracted in a predetermined order and includes a phrase translation unit 221, a user word translation unit 222, a learning word translation unit 223, an abbreviation translation unit 224 ), A vocabulary translation unit 225, a word translation unit 226, and a prediction translation unit 227.

The phrase translation unit 221 is a unit for translating a phrase extracted by the phrase extraction unit 211. [ And a Hangul phrase corresponding to the extracted phrase is retrieved from the phrase DB 350. [

The user word translation unit 222 is a unit for translating the user word extracted by the user word extraction unit 212. And the Korean word corresponding to the extracted user word is loaded from the user word DB 320. [

The learning word translation unit 223 is a unit for translating the learning word extracted by the learning word extraction unit 213. [ And the Hangul words corresponding to the extracted learning words are retrieved from the learning word DB 330. [

The abbreviation-translating unit 224 is a unit for translating the abbreviation extracted by the abbreviation-extracting unit 214. And a Hangul word corresponding to the extracted abbreviation is loaded from the abbreviation DB 340.

The vocabulary translation unit 225 is a unit for translating the vocabulary extracted by the vocabulary extraction unit 215. And the Korean language vocabulary corresponding to the extracted vocabulary is retrieved from the English dictionary DB 310.

The word translation unit 226 is a unit for translating the word extracted by the word extraction unit 216. [ And the Korean word corresponding to the extracted word is retrieved from the English dictionary DB 310. [

The predictive translation unit 227 is a unit for translating the word extracted by the non-registered extraction unit 217. Since the extracted word is a word not stored in the database 300, it can be output as it is without translating it, and can be input by a user, or translated into a selected Korean word by searching the web.

At this time, the word received by the user is stored in the user word DB 320, and the word retrieved from the web is stored in the learning word DB 330.

The translated-text output unit 230 includes a unit sentence output unit 231, a detailed unit output unit 232, and a translation information display unit 233.

The unit sentence output unit 231 outputs a sentence sentence 231a extracted by the extracting unit 210 as a unit sentence and a translated sentence 231b corresponding thereto. The original sentence 231a is a sentence containing English extracted as a unit sentence in the extraction unit 210 and the translated sentence 231b is a combination of the translated words 232b translated into Korean in the detailed unit output unit 232 .

The detail unit output unit 232 is a unit for outputting a source word 232a dividing the sentence sentence 231a into detailed translation units and a translated word 232b corresponding thereto. A detailed translation unit may be the above-mentioned phrase, user word, learning word, abbreviation, vocabulary, and if not, a word separated by a space or a punctuation code becomes a detailed translation unit. Translation is carried out by detailed translation unit. That is, a Hangul word corresponding to each phrase, a user word, a learning word, an abbreviation, a vocabulary, and the like is retrieved from the database 300 and matched with 1: 1. In this case, as shown in FIG. 1, the corresponding original word 232a and the translated word 232b can be located in the same segment, and the corresponding relationship can be grasped at a glance. This makes it easy to check whether or not the translation has been correctly translated and improve the translation accuracy. In addition, the translation word (232b) may be automatically translated into a different color according to a word that is automatically translated, such as an automatically translated word, a user word, a learning word, a word that is not translated, It is easy to grasp the translation state by reflecting it in the translation sentence 231b. That is, when the translated word is displayed in red and the abbreviation is displayed in blue as shown in FIG. 1 and reflected in the translation sentence 231b, information such as which word is translated and which word is abbreviated can be confirmed at a glance It is.

The detailed unit output unit 232 may further include a translated word correcting unit 232c and a user's word input unit 232d.

The translated word correcting unit 232c is a unit for correcting the translated word 232b into another translated word 232b 'for each unit of the detailed unit output unit 232. The original word 232a and the translated word 232b It is used when you want to output a meaning other than the translated Korean word when checking which English word is translated into which Korean word by placing it in the same division box. The user may directly input a new Korean word meaning by pressing the translated word correcting unit 232c or may present a similar Korean word list in the database 300 for the user to select. At this time, the selected Hangul words are stored in the learning word DB 330 together with the corresponding English words.

The user word input unit 232d is a unit for inputting the original word 232a of the detailed detailed unit output unit 232 as a user word and the original word 232a and the translation word 232b are located in the same division cell, It is used when confirming the English word is translated into a Korean word, and when it is desired to directly translate a word remaining in English without being translated. It is also possible to directly input a new Korean word meaning by pressing the user word input section 232d as a word which is not stored in the database 300 or which has not been translated by the predictive translation section 227 or may be inputted to the predictive translation section 227 A list of Korean words found by the user can be presented to the user. At this time, the selected Hangul word is stored in the user word DB 320 together with the corresponding English word.

The translation information display unit 233 displays translation information including the number of detailed translation units, the number of non-target words, the number of unregistered words, and the translation ratio. The detailed translation unit includes a phrase extraction unit 211, a user word extraction unit 212, a learning word extraction unit 213, an abbreviation extraction unit 214, a lexical extraction unit 215 or the word extracting unit 216, respectively. The number of non-target words means the number of sentence codes or Korean words extracted from the non-target extracting unit 218 in the unit sentence extracted by the extracting unit 210. The number of unregistered words means the number of words extracted by the unregistered extracting unit 217 in the unit sentence extracted by the extracting unit 210. The translation rate can be calculated by comparing the number of words to be translated and the number of words actually translated in the translation word 232b in the original word 232a and displayed on the translation information display unit 233. [

The translation creating module 400 is a part for creating the medical recording translation 30 by reflecting the translated result by the medical recording translation module 200 on the medical recording text text 20. After the translation work is completed on the screen as shown in FIG. 1, the medical record translated book 30 is created by substituting the sentence sentence 231a at each corresponding position of the medical record paper text book 20 as shown in FIG. The medical record translation 30 created at this time can be seen in FIG. Preferably, as shown in FIG. 8, the original medical document and the translated document may be output on one screen so that the corresponding relationship can be confirmed at a glance. In particular, since the position of each phrase in the original and the translation is matched, a translation corresponding to the original English text on the original can be confirmed at the same position of the translation, and conversely, the original text corresponding to the translation Therefore, ambiguous meaning can be grasped easily. Therefore, when the screen as shown in FIG. 8 is presented, it becomes much easier to create a report based on the screen.

The database 300 includes an English dictionary DB 310, a user word DB 320, a learning word DB 330, and an abbreviation dictionary DB 340, where data to be translated is stored.

The English dictionary DB 310 is a database in which English words including English medical terms and corresponding Korean words are stored.

The user word DB 320 is a database in which user words and corresponding Korean words are stored.

The learning word DB 330 is a database in which learning words and corresponding Korean words are stored.

The abbreviation dictionary DB (340) is a database storing abbreviated English medical terms and their corresponding Hangul words.

The phrase DB 350 is a database for storing English medical terms and phrases corresponding to phrases.

Since the data stored in the database is stored in the server, the user can refer to user words, abbreviations, and phrases registered by other employees when the loss assessment company shares them (refer to FIG. 4).

A method of translating a medical recording sheet by the above-described configuration is performed in the following order.

s1: extracting text from the medical recording paper scanned book 10 and generating a medical recording paper text book 20

s1-1: The server receives the medical record scanned book 10 and converts it into a file format capable of extracting text.

s1-2: extracts the text by recognizing the image in the formatted file of s1-1.

s1-3: The text extracted from s1-2 is stored as a medical recording paper text 20.

s2: translating the English text of the medical recording paper text 20 into Korean

s2-1: Abbreviation of the text of the medical record paper text (20) is extracted and translated. Medical terminology is most abbreviated, so abbreviation translation is the top priority.

s2-2: The text of the medical record paper text 20 is extracted and translated on a sentence-by-sentence basis. At this time, a sentence unit separated by a punctuation mark such as a period is extracted including the abbreviation translated in s2-1. If the entire extracted sentence is not translated, the translated phrases (registered in the database) are extracted and translated by reducing the length of the phrases sequentially by excluding the last word of the sentence. When a part of the front part of a sentence is extracted as a phrase, the same operation is performed in the remaining part to extract and translate the phrase.

s2-3: extract the user words from the untranslated texts in s2-1 to s2-2 and translate them.

s2-4: s2-1 to s2-3 extract the learning word from the untranslated text and translate it.

s2-5: s2-1 to s2-4 extract vocabulary from untranslated text and translate it.

s2-6: extracts the registered word from the untranslated text in s2-1 to s2-5 and translates it.

s2-7: Extracts unregistered words and non-word parts and Hangul characters from untranslated text in s2-1 to s2-6. In this case, since the extracted text is a part that can not be automatically translated, it is output as it is in s2-8.

s2-8: The original sentence extracted from s2-2, the original word extracted from s2-1 ~ s2-7, the translated word translated from s2-1 ~ s2-6, .

s2-9: Input the Korean word corresponding to the unregistered word of s2-7 directly or search the web and enter it.

s3: a step of creating a medical recording translated book 30 reflecting the translated result

s3-1: Creates the medical recording translation 30 by replacing the translated sentence in which the translation work has been completed in the s2 with the original sentence text of the medical recording paper text 20.

s3-2: The medical record book scan book 10 or the medical record book text book 20 and the medical book translation book 30 are displayed on a screen such that the text positions of the original text and the translated text correspond to each other as shown in FIG.

While the present invention has been described in connection with what is presently considered to be practical exemplary embodiments, it is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, It will be clear to those who have knowledge.

100: medical record scanning module
110: scan file receiver
120: File format conversion unit
130:
140: Text file generation unit
200: medical record translation module
210:
211:
212: user word extracting unit
213: Learning word extraction unit
214: Abbreviation extraction unit
215:
216: registered word extracting unit
217: Unregistered Extraction Unit
218: Non-target extraction unit
219: Personal Information Extraction Unit
220: translation unit
221: Translation section
222: user word translation unit
223: Learning word translation part
224: Abbreviation translation part
225: Vocabulary Translation Department
226: Word translation part
227: Prediction translation unit
230: translation output unit
231: unit sentence output unit
231a: One sentence
231b: translation sentence
232: detailed unit output section
232a: One word
232b: Translation Words
232c: translation word counting
232d: user word input section
233: Translation information display section
300: Database
310: English dictionary DB
320: User Word DB
330: Learning Word DB
340: Abbreviation Dictionary DB
350: passage DB
400: Translation module
10: Scanned medical records
20: medical record text text
30: Translation of medical records

Claims (4)

A medical record scanning module (100) for converting the medical record scan book (10) into a medical record text book (20);
A medical recording paper translation module 200 for translating the English text of the medical recording paper text 20 into Korean;
A database 300 in which data to be translated is stored;
And a translation creating module (400) for creating a translated medical record translation book (30) by reflecting the translated result by the medical record translation module (200) to the medical record text book (20)

The database (300)
And a phrase database 350 in which English medical terms in phrases and corresponding Hangul phrases are stored,

The medical record translation module 200
An extracting unit 210 for extracting a translation unit, which is a unit sentence separated by punctuation marks or line breaks, in the medical recording paper text 20,
A translator 220 for translating the extracted translation units in a predetermined order,
And a translation output unit (230) for outputting translation results in translation units,

The extracting unit 210 extracts
An abbreviation extracting unit 214 for extracting abbreviations from unit sentences,
A phrase extracting unit 211 for extracting the longest translatable phrase in the unit sentence,
A user word extracting unit 212 for extracting a user word among texts not extracted by the abbreviation extracting unit 214 and the phrase extracting unit 211 in a unit sentence,
An abbreviation extracting unit 214 and a phrase extracting unit 211 in a unit sentence, a learning word extracting unit 213 extracting a learning word among texts not extracted by the user word extracting unit 212,
A vocabulary extracting unit 215 for extracting vocabulary among the texts not extracted by the abbreviation extracting unit 214 and the phrase extracting unit 211, the user word extracting unit 212 and the learning word extracting unit 213,
Extracts non-word (punctuation marks, etc.) or Hangul from non-extracted text in the abbreviation extracting unit 214, the phrase extracting unit 211, the user word extracting unit 212, and the learning word extracting unit 213, The target extraction unit 218,
And a personal information extracting unit (219) for extracting personal information from the text of the translation target file,
The user word is an English word registered by the user,
The learning word is characterized in that the translated word is an English word modified by the user,

The phrase extraction unit 211 extracts
If the whole unit sentence is stored in the phrase database 350, the sentence is extracted as a phrase. Otherwise, the phrase stored in the phrase database 350 is extracted while sequentially reducing the length of the phrase by excluding the last word,
If one phrase is extracted, if the remaining words are stored in the phrase database 350 while sequentially decreasing the length of the phrase, the extracted phrase is extracted as a phrase,

The translator 220 translates
An abbreviation translating unit 224 for translating abbreviations extracted by the abbreviation extracting unit 214,
A phrase translation unit 221 for translating a phrase extracted by the phrase extraction unit 211,
A user word translation unit 222 for translating the user word extracted by the user word extraction unit 212,
A learning word translation unit 223 for translating the learning word extracted by the learning word extraction unit 213,
A vocabulary translation unit 225 for translating the vocabulary extracted by the vocabulary extraction unit 215,
And a word translation unit 226 for translating the word extracted by the word extraction unit 216,

The translated text output unit 230
A unit sentence output unit 231 for outputting an original sentence 231a extracted as a unit sentence by the extraction unit 210 and a translated sentence 231b corresponding thereto,
A detailed unit output unit 232 for outputting a source word 232a dividing the sentence sentence 231a into detailed translation units and a translation word 232b corresponding to the source word 232a in the same segment,
A translation information display unit 233 for displaying translation information including the number of detailed translation units, the number of non-target words, the number of unregistered words, and the translation ratio,
A detailed translation unit is one of a phrase, a user word, a learning word, an abbreviation, and a vocabulary, and a translation is performed for each detailed translation unit,
(232b) for each detailed translation unit, an automatically translated word including a vocabulary and a user's word, a manually translated word including a learned word, a non-translated word, and a non-translated word Color, and reflects it in the translation sentence 231b,

And the translation writing module (400) substitutes the original sentence (231a) at each corresponding position of the medical recording paper text book (20) to create the medical recording translation book (30)
Medical record translation system.

delete delete delete

Images