Fast and interpretable mortality risk scores for critical care patients.

Journal: Journal of the American Medical Informatics Association : JAMIA
PMID:

Abstract

OBJECTIVE: Prediction of mortality in intensive care unit (ICU) patients typically relies on black box models (that are unacceptable for use in hospitals) or hand-tuned interpretable models (that might lead to the loss in performance). We aim to bridge the gap between these 2 categories by building on modern interpretable machine learning (ML) techniques to design interpretable mortality risk scores that are as accurate as black boxes.

Authors

  • Chloe Qinyu Zhu
    Department of Computer Science, Duke University, Durham, NC 27708, United States.
  • Muhang Tian
    Department of Computer Science, Duke University, Durham, NC 27708, United States.
  • Lesia Semenova
    Microsoft Research, Duke University, Durham, United States.
  • Jiachang Liu
    Duke University.
  • Jack Xu
    Department of Computer Science, Duke University, Durham, NC 27708, United States.
  • Joseph Scarpa
    Anesthesiology, Weill Cornell Medicine.
  • Cynthia Rudin
    Duke University.