Automatic ICD-10 Code Association: A Challenging Task on French Clinical Texts

TL;DR

This paper presents a novel NLP-based model for automatically assigning ICD-10 codes to French clinical texts, significantly improving accuracy over previous methods by leveraging recent advances in transformer models and multi-label classification.

Contribution

It introduces a new model combining recent NLP techniques with multi-label classification tailored for French clinical texts, achieving substantial performance gains.

Findings

F1-score increased by over 55% compared to previous state-of-the-art.

Model effectively handles large input and label sets in French medical texts.

Demonstrates the applicability of transformer-based models in non-English clinical NLP tasks.

Abstract

Automatically associating ICD codes with electronic health data is a well-known NLP task in medical research. NLP has evolved significantly in recent years with the emergence of pre-trained language models based on Transformers architecture, mainly in the English language. This paper adapts these models to automatically associate the ICD codes. Several neural network architectures have been experimented with to address the challenges of dealing with a large set of both input tokens and labels to be guessed. In this paper, we propose a model that combines the latest advances in NLP and multi-label classification for ICD-10 code association. Fair experiments on a Clinical dataset in the French language show that our approach increases the $F_1$-score metric by more than 55\% compared to state-of-the-art results.