Medical named entity identification method utilizing multi-model fusion strategy

Abstract

The invention relates to a medical named entity identification method utilizing a multi-model fusion strategy, which is technically characterized by comprising the following steps of: manually markingmedical concepts appearing in a text to obtain a training text set; preprocessing the training text set data to obtain texts after word segmentation; constructing word features, part-of-speech features and medical features of words in the text after word segmentation to obtain complete coding features of the words; respectively inputting the text after word segmentation and the complete coding characteristics of the words into odd number of sequence labeling models for model learning to obtain corresponding model parameters; and fusing the preliminary annotation results according to a manually specified rule to obtain a final sequence annotation result. According to the method, the accuracy and recall ratio of medical entity automatic labeling are improved by effectively utilizing model diversity and nonlinear modeling capability, the method can be widely applied to medical entity labeling work of a non-numerical class, a new guide is provided for medical named entity labeling research, and the method has remarkable significance in promoting intelligent medical treatment.

Description

technical field [0001] The invention belongs to the technical field of medical information data processing, in particular to a medical named entity recognition method using a multi-model fusion strategy. Background technique [0002] Medical records are the records of the medical process such as inspection, diagnosis and treatment by medical staff on the occurrence, development and outcome of patients' diseases. health records. Medical records play an important role in medical treatment, prevention, teaching, scientific research, and hospital management. With the development of Internet technology, most hospitals have realized the electronicization of clinical medical records. Electronic medical records are digital patient medical records that are stored, managed, transmitted and reproduced with electronic equipment, replacing handwritten paper medical records. They are proactive, complete, and Accurate, knowledge related, timely acquisition and other advantages. [0003]...

AI Technical Summary

Problems solved by technology

At present, the commonly used medical named entity recognition in electronic medical records mainly adopts a single-model strategy. However, this method has limited accuracy and recall and is cumbersome and complicated.

Images