Accuracy Analysis of Triage Recommendation Based on CNN, RNN and RCNN Models

摘要

It is an universal phenomenon for patients who do not know which clinical department to register in large general hospitals. Although triage nurses can help patients, due to the lager number of patients, they have to stand in a queue for minutes to consult. In order to solve these problems, we propose to use deep learning models for recommending clinical departments automatically, which help patients to complete self-diagnosis, improve the efficiency of triage, reduce the work intensity of triage nurses. In this study, we construct a Chinese corpus with 20,000 texts of patient symptom descriptions and use word2vec to generated word vector. In addition, we analyze the performance of CNN, RNN and RCNN in clinical departments recommendation. The results of our experiment shows that the accuracy of RCNN is improved markedly than CNN and RNN. Therefore, RCNN is more suitable for the task of clinical departments recommendation.