Quantifying surprise in clinical care: Detecting highly informative events in electronic health records with foundation models
By: Michael C. Burkhart , Bashar Ramadan , Luke Solo and more
Potential Business Impact:
Finds important patient health clues in records.
We present a foundation model-derived method to identify highly informative tokens and events in electronic health records. Our approach considers incoming data in the entire context of a patient's hospitalization and so can flag anomalous events that rule-based approaches would consider within a normal range. We demonstrate that the events our model flags are significant for predicting downstream patient outcomes and that a fraction of events identified as carrying little information can safely be dropped. Additionally, we show how informativeness can help interpret the predictions of prognostic models trained on foundation model-derived representations.
Similar Papers
Foundation Models for Clinical Records at Health System Scale
Machine Learning (CS)
Predicts future illnesses from patient records.
FoMoH: A clinically meaningful foundation model evaluation for structured electronic health records
Machine Learning (CS)
Helps doctors predict sickness using patient records.
Foundation models for electronic health records: representation dynamics and transferability
Machine Learning (CS)
Helps doctors predict patient health using shared data.