Revisiting the MIMIC-IV Benchmark: Experiments Using Language Models for Electronic Health Records
This work addresses a barrier for adopting language models in medical applications, but it is incremental as it revisits an existing benchmark with standard methods.
The paper tackled the lack of standardized benchmarks for natural language models in healthcare by revisiting the MIMIC-IV dataset for electronic health records, showing that fine-tuned text-based models are competitive against tabular classifiers on a mortality prediction task, while zero-shot LLMs performed poorly.
The lack of standardized evaluation benchmarks in the medical domain for text inputs can be a barrier to widely adopting and leveraging the potential of natural language models for health-related downstream tasks. This paper revisited an openly available MIMIC-IV benchmark for electronic health records (EHRs) to address this issue. First, we integrate the MIMIC-IV data within the Hugging Face datasets library to allow an easy share and use of this collection. Second, we investigate the application of templates to convert EHR tabular data to text. Experiments using fine-tuned and zero-shot LLMs on the mortality of patients task show that fine-tuned text-based models are competitive against robust tabular classifiers. In contrast, zero-shot LLMs struggle to leverage EHR representations. This study underlines the potential of text-based approaches in the medical field and highlights areas for further improvement.