MDACE: MIMIC Documents Annotated with Code Evidence
Authors:
- Hua ChengHCHua Cheng
- Rana JafariRJRana Jafari
- April RussellARApril Russell
- Russell KlopferRKRussell Klopfer
- Edmond LuELEdmond Lu
- Benjamin StrinerBSBenjamin Striner
- Matthew R. GormleyMGMatthew R. Gormley
Preprint
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.Keywords
All Related Versions
This publication has 15 references indexed in Scilit:
- A Unified Review of Deep Learning for Automated Medical CodingACM Computing Surveys, 2024
- MIMIC-IV, a freely accessible electronic health record datasetScientific Data, 2023
- Hierarchical label-wise attention transformer model for explainable ICD codingJournal of Biomedical Informatics, 2022
- An explainable CNN approach for medical codes prediction from clinical textBMC Medical Informatics and Decision Making, 2021
- Effective Convolutional Attention Network for Multi-label Clinical Document Classification2021
- Thermostat: A Large Collection of NLP Model Explanations and Analysis Tools2021
- Weakly- and Semi-supervised Evidence Extraction2020
- A Simple and Effective Model for Answering Multi-span Questions2020
- Experimental Evaluation and Development of a Silver-Standard for the MIMIC-III Clinical Coding Dataset2020
- EHR Coding with Multi-scale Feature Attention and Structured Knowledge Graph Propagation2019