Towards BERT-based Automatic ICD Coding: Limitations and Opportunities
This addresses the challenge of ICD coding for medical applications, but it is incremental as it identifies limitations and opportunities without major breakthroughs.
The paper tackled the problem of automatic ICD coding from medical notes using BERT-based models, finding that fine-tuning on long texts is a key limitation, and showed that pretrained transformers can achieve competitive performance with small text portions.
Automatic ICD coding is the task of assigning codes from the International Classification of Diseases (ICD) to medical notes. These codes describe the state of the patient and have multiple applications, e.g., computer-assisted diagnosis or epidemiological studies. ICD coding is a challenging task due to the complexity and length of medical notes. Unlike the general trend in language processing, no transformer model has been reported to reach high performance on this task. Here, we investigate in detail ICD coding using PubMedBERT, a state-of-the-art transformer model for biomedical language understanding. We find that the difficulty of fine-tuning the model on long pieces of text is the main limitation for BERT-based models on ICD coding. We run extensive experiments and show that despite the gap with current state-of-the-art, pretrained transformers can reach competitive performance using relatively small portions of text. We point at better methods to aggregate information from long texts as the main need for improving BERT-based ICD coding.