Code Synonyms Do Matter: Multiple Synonyms Matching Network for Automatic ICD Coding

TL;DR

This paper introduces a novel multiple synonyms matching network for automatic ICD coding, leveraging code synonyms aligned with UMLS to improve code representation and classification accuracy in EMRs.

Contribution

The paper proposes a new method that incorporates multiple code synonyms aligned with UMLS, enhancing code representations for better ICD code prediction.

Findings

Outperforms previous state-of-the-art methods on MIMIC-III dataset

Utilizes code synonyms to improve code representation learning

Demonstrates the effectiveness of synonym-based matching in ICD coding

Abstract

Automatic ICD coding is defined as assigning disease codes to electronic medical records (EMRs). Existing methods usually apply label attention with code representations to match related text snippets. Unlike these works that model the label with the code hierarchy or description, we argue that the code synonyms can provide more comprehensive knowledge based on the observation that the code expressions in EMRs vary from their descriptions in ICD. By aligning codes to concepts in UMLS, we collect synonyms of every code. Then, we propose a multiple synonyms matching network to leverage synonyms for better code representation learning, and finally help the code classification. Experiments on the MIMIC-III dataset show that our proposed method outperforms previous state-of-the-art methods.