Fusion Model for Tentative Diagnosis Inference Based on Clinical Narratives

• DOI: 10.26599/tst.2022.9010049
• Publication Date: 2023-01-06T18:48:04Z
• Authors: Ying Yu; Junwen Duan; Min Li
• Keywords: Economic shortage; Code (set theory); Value (mathematics)

In general, physicians make a preliminary diagnosis based on patients' admission narratives and conditions, largely depending their experiences professional knowledge. An automatic accurate tentative clinical would be of great importance to physicians, particularly in the shortage medical resources. Despite its value, little work has been conducted this method. Thus, study, we propose fusion model that integrates semantic symptom features contained text. The input text are initially captured by an attention-based Bidirectional Long Short-Term Memory (BiLSTM) network. concepts, recognized from text, then vectorized using term frequency-inverse document frequency method relations between symptoms diseases. Finally, two strategies utilized recommend most potential candidate for international classification diseases code. Model training evaluation performed public dataset. results show both achieved promising performance, which best performance obtained top-3 accuracy 0.7412.