A method for converting a medical record text into structured metadata from a natural language

Abstract

The invention relates to a method for converting a medical record text into structured metadata from a natural language, which comprises the following steps of: 1, extracting characteristic value texts of specific organs, parts and surgical expressions in a text format from a historical case report, and carrying out characteristic value analysis on the characteristic value texts to obtain a characteristic dictionary; 2, exporting historical detection reports needing to be analyzed from a hospital, and merging the historical detection reports into a to-be-processed data set; Step 3, traversingthe patient cases in the data set, segmenting words according to the characteristic value dictionary, and intercepting the description of the organ, the part or the surgical expression; and step 4, persisting the data content intercepted by the part into a structured database. According to the method, historical cases are utilized, a traversal algorithm is used, the use cost is reduced, the generation step of a training set is omitted, lesions of different organs and parts can be conveniently counted, analyzed and searched for through the converted structured cases, and doctors can conveniently carry out medical research, paper writing and teaching.

Description

technical field [0001] The invention belongs to the field of medical information, and in particular relates to a method for converting medical record text from natural language into structured metadata. Background technique [0002] With the development of information technology and the construction of medical informatization, hospitals are now gradually using structured reporting systems for report writing, but when outputting to printed reports, the information display forms such as image performance and diagnosis are single, basically The form of label and value, or the pattern with number and value, obviously this form cannot be applied to all types of reports, and domestic hospitals and various system requirements require text reports, so doctors need to write a new one report, or modify the report form of labels and values ​​to generate our natural language report, which increases the workload of doctors and reduces efficiency. [0003] The patent with the application...

AI Technical Summary

Problems solved by technology

[0002] With the development of information technology and the construction of medical informatization, hospitals are now gradually using structured reporting systems for report writing, but when outputting to printed reports, the information display forms such as image performance and diagnosis are single, basically The form of label and value, or the pattern with number and value, obviously this form cannot be applied to all types of reports, and domestic hospitals and various system requirements require text reports, so doctors need to write a new one report, or modify the report form of labels and values ​​to generate our natural language report, which increases the workload of doctors and reduces efficiency

[0003] The patent with the application number "CN201810375610.8" titled "A Method and System for Generating Medical Reports from Pre-structured Data" uses encapsulation modules to replace text to achieve the purpose of generating different structured reports. The text descriptions of each field in the field cannot be individually edited into doctor's idiomatic sentences, which still does not conform to the usage habits for the doctor's experience

Images