Foundation models for electronic health records: representation dynamics and transferability
This work addresses the challenge of adapting foundation models to local health systems with limited data, which is incremental as it builds on existing FM approaches for EHRs.
The study investigated the transferability of a foundation model trained on MIMIC-IV EHR data to a local dataset from the University of Chicago Medical Center, assessing outlier detection and patient trajectories, and found insights into adaptability and predictive factors across healthcare systems.
Foundation models (FMs) trained on electronic health records (EHRs) have shown strong performance on a range of clinical prediction tasks. However, adapting these models to local health systems remains challenging due to limited data availability and resource constraints. In this study, we investigated what these models learn and evaluated the transferability of an FM trained on MIMIC-IV to an institutional EHR dataset at the University of Chicago Medical Center. We assessed their ability to identify outlier patients and examined representation-space patient trajectories in relation to future clinical outcomes. We also evaluated the performance of supervised fine-tuned classifiers on both source and target datasets. Our findings offer insights into the adaptability of FMs across different healthcare systems, highlight considerations for their effective implementation, and provide an empirical analysis of the underlying factors that contribute to their predictive performance.