A medical record text named entity recognition method based on an iterative expansion convolutional neural network

Abstract

The invention provides a medical record text named entity recognition method based on an iterative expansion convolutional neural network. According to the method, the named entity recognition is carried out on a medical electronic medical record data set CCKS2017, a section of Chinese electronic medical record text is inputted, an iterative expansion convolutional neural network and a conditionalrandom field are used as a model architecture, the Chinese character components are used as features, and the named entities such as disease names and inspection means in the text are extracted.

Description

technical field [0001] The present invention relates to natural language processing and medical related fields, more specifically, to a named entity recognition method for medical record text based on iteratively expanded convolutional neural network. Background technique [0002] In recent years, with the development of big data and computer technology, more and more medical institutions have begun to adopt electronic medical record systems. The electronic medical record system is a special software for medicine. The hospital electronically records the patient's medical information through electronic medical records, including: medical history, medical course records, inspection and test results, doctor's orders, surgical records, nursing records, etc., including both structured information and unstructured free text. There is also graphic image information. [0003] With the development of artificial intelligence technology, many teams have begun to try to use artificial...

AI Technical Summary

Problems solved by technology

[0005] The named entity recognition in the medical field is different from the general field. The main differences are as follows: (1) There are many professional terms and rare words in the medical field, such as "loratadine tablets". The current Chinese word segmentation tools cannot perform word segmentation well. Therefore, it will affect the subsequent recognition effect

(2) Some entity names are long, such as "brain protein hydrolyzate nourishes brain cells" (treatment), and some models are difficult to establish a long context dependence

[0006] For the first question, considering that the existing word segmentation tools have a poor word segmentation effect on medical texts, here we no longer perform word segmentation and directly operate on Chinese characters

Images