Example Code to accompany paper: Machine Learning for Patient Risk Stratification: Standing on, or looking over, the shoulders of clinicians?

Link: @TODO.

DOI: @TODO.

Authors: Brett K. Beaulieu-Jones*, William Yuan*, Gabriel A. Brat, Andrew L. Beam, Griffin Weber, Marshall Ruffin, Isaac S. Kohane

Abstract: Machine learning can help clinicians to make individualized patient predictions only if researchers demonstrate models that contribute novel insights, rather than learning from actions clinicians have already taken. We trained deep learning models using only clinician-initiated, administrative data for 42.9 million admissions using three subsets of data: demographic data only, demographic data and information available at admission, and the previous data plus charges recorded during the first day of admission. Models trained on charges during the first day of admission achieve performance close to published EMR-based benchmarks for inpatient outcomes: in-hospital mortality (0.89 AUC), prolonged length of stay (0.82 AUC) and 30-day readmission rate (0.71 AUC). Furthermore, these models exhibited significant declines in performance when evaluated over only myocardial infarction (MI) patients relative to models trained over MI patients alone, highlighting the importance of physician diagnosis in the prognostic performance of these models. These results provide a benchmark for predictive accuracy trained only on prior clinical actions and indicate that models with similar performance may derive their signal by looking over clinician's shoulders--using clinical behavior as the expression of pre-existing intuition and suspicion to generate a prediction.