Impact of Longitudinal Trends in Predicting Pediatric Clinical Deterioration

Poster Number: P24
Presentation Time: 05:00 PM - 06:30 PM

Abstract Keywords: Machine Learning, Deep Learning, Pediatrics, Patient Safety
Primary Track: Applications
Programmatic Theme: Clinical Informatics

Clinical deterioration among pediatric patients is associated with poor health outcomes. We trained LSTM models that predict which pediatric patients are at an imminent risk of experiencing clinical deterioration. The performance of our constructed LSTM models is compared with that achieved by our previously published gradient boosted machine models, called pCART, on the same longitudinal datasets. While our LSTM models did not outperform pCART, training an LSTM on pCART outputs lead to more generalizable performance.

Speaker(s):
Sierra Strutz, PhD Student in Biomedical Data Science
University of Wisconsin - Madison

Author(s):
Sierra Strutz, PhD Student in Biomedical Data Science - University of Wisconsin - Madison; Kyle Carey, MPH - University of Chicago; Fereshteh S. Bashiri, PhD - University of Wisconsin - Madison; Priti Jani, MD, MPH - University of Chicago; Emily Gilbert, MD - Loyola University; Majid Afshar, MD, MSCR - University of Wisconsin - Madison; Matthew Churpek, MD, MPH, PhD - University of Wisconsin-Madison; Anoop Mayampurath, PhD - University of Wisconsin - Madison;