Foundation model of electronic medical records for adaptive risk estimation

Hospitals struggle to predict critical outcomes. Traditional early warning systems, like NEWS and MEWS, rely on static variables and fixed thresholds, limiting their adaptability, accuracy, and personalization.

We previously developed the Enhanced Transformer for Health Outcome Simulation (ETHOS), an artificial intelligence (AI) model that tokenizes patient health timelines (PHTs) from electronic health records and uses transformer-based architectures to predict future PHTs. ETHOS is a versatile framework for developing a wide range of applications. In this work, we develop the Adaptive Risk Estimation System (ARES) that leverages ETHOS to compute dynamic, personalized risk probabilities for clinician-defined critical events. ARES also features a personalized explainability module that highlights key clinical factors influencing risk estimates. We evaluated ARES using the MIMIC-IV v2.2 dataset, together with its emergency department extension, and benchmarked performance against both classical early warning systems and contemporary machine learning models.

The entire dataset was tokenized, resulting in 285,622 PHTs (63% with at least 1 hospital admission), comprising over 357 million tokens. ETHOS outperformed benchmark models in predicting hospital admissions, intensive care unit admissions, and prolonged stays, achieving superior area under the curve scores. Its risk estimates were robust across demographic subgroups, with calibration curves confirming model reliability. The explainability module provided valuable insights into patient-specific risk factors.

ARES, powered by ETHOS, advances predictive health care AI by delivering dynamic, real-time, personalized risk estimation with patient-specific explainability. Although our results are promising, the clinical impact remains uncertain. Demonstrating ARES's true utility in real-world settings will be the focus of our future work.