Medical Code Prediction from Discharge Summary: Document to Sequence BERT using Sequence Attention

TL;DR

This paper introduces a BERT-based model with sequence attention for automatic ICD code prediction from clinical notes, achieving superior performance on the MIMIC-III dataset.

Contribution

It presents a novel BERT-based approach with sequence attention for document classification, specifically for ICD code prediction from discharge summaries.

Findings

Achieved macro-averaged F1 of 0.629 and micro-averaged F1 of 0.686 on MIMIC-III.

Outperformed existing state-of-the-art models on the same dataset.

Demonstrated effectiveness of sequence attention in capturing important information in clinical documents.

Abstract

Clinical notes are unstructured text generated by clinicians during patient encounters. Clinical notes are usually accompanied by a set of metadata codes from the International Classification of Diseases(ICD). ICD code is an important code used in various operations, including insurance, reimbursement, medical diagnosis, etc. Therefore, it is important to classify ICD codes quickly and accurately. However, annotating these codes is costly and time-consuming. So we propose a model based on bidirectional encoder representations from transformers (BERT) using the sequence attention method for automatic ICD code assignment. We evaluate our approach on the medical information mart for intensive care III (MIMIC-III) benchmark dataset. Our model achieved performance of macro-averaged F1: 0.62898 and micro-averaged F1: 0.68555 and is performing better than a performance of the state-of-the-art model using the MIMIC-III dataset. The contribution of this study proposes a method of using BERT that can be applied to documents and a sequence attention method that can capture important sequence in-formation appearing in documents.