Automated Medical Coding on MIMIC-III and MIMIC-IV: A Critical Review and Replicability Study
This work addresses the problem of unreliable automated medical coding evaluations for healthcare professionals, but it is incremental as it focuses on replication and correction rather than new methods.
The paper reproduces and analyzes state-of-the-art automated medical coding models on MIMIC-III and MIMIC-IV datasets, revealing that previous models underperform due to methodological flaws, with a correction doubling the macro F1 score, and provides the first comprehensive results on MIMIC-IV.
Medical coding is the task of assigning medical codes to clinical free-text documentation. Healthcare professionals manually assign such codes to track patient diagnoses and treatments. Automated medical coding can considerably alleviate this administrative burden. In this paper, we reproduce, compare, and analyze state-of-the-art automated medical coding machine learning models. We show that several models underperform due to weak configurations, poorly sampled train-test splits, and insufficient evaluation. In previous work, the macro F1 score has been calculated sub-optimally, and our correction doubles it. We contribute a revised model comparison using stratified sampling and identical experimental setups, including hyperparameters and decision boundary tuning. We analyze prediction errors to validate and falsify assumptions of previous works. The analysis confirms that all models struggle with rare codes, while long documents only have a negligible impact. Finally, we present the first comprehensive results on the newly released MIMIC-IV dataset using the reproduced models. We release our code, model parameters, and new MIMIC-III and MIMIC-IV training and evaluation pipelines to accommodate fair future comparisons.