MDACE: MIMIC Documents Annotated with Code Evidence

TL;DR

This paper introduces MDACE, a new annotated dataset of medical documents with evidence spans for coding, enabling improved evaluation of evidence extraction methods in clinical coding systems.

Contribution

The paper presents MDACE, the first publicly available dataset with expert-annotated evidence spans for medical coding, facilitating research in evidence extraction and model interpretability.

Findings

Baseline evidence extraction methods achieve moderate performance.

MDACE enables evaluation of deep learning models for evidence extraction.

The dataset supports research in explainability for clinical coding.

Abstract

We introduce a dataset for evidence/rationale extraction on an extreme multi-label classification task over long medical documents. One such task is Computer-Assisted Coding (CAC) which has improved significantly in recent years, thanks to advances in machine learning technologies. Yet simply predicting a set of final codes for a patient encounter is insufficient as CAC systems are required to provide supporting textual evidence to justify the billing codes. A model able to produce accurate and reliable supporting evidence for each code would be a tremendous benefit. However, a human annotated code evidence corpus is extremely difficult to create because it requires specialized knowledge. In this paper, we introduce MDACE, the first publicly available code evidence dataset, which is built on a subset of the MIMIC-III clinical records. The dataset -- annotated by professional medical coders -- consists of 302 Inpatient charts with 3,934 evidence spans and 52 Profee charts with 5,563 evidence spans. We implemented several evidence extraction methods based on the EffectiveCAN model (Liu et al., 2021) to establish baseline performance on this dataset. MDACE can be used to evaluate code evidence extraction methods for CAC systems, as well as the accuracy and interpretability of deep learning models for multi-label classification. We believe that the release of MDACE will greatly improve the understanding and application of deep learning technologies for medical coding and document classification.