Predicting sepsis with a recurrent neural network using the MIMIC III database.

Journal: Computers in biology and medicine
Published Date:

Abstract

OBJECTIVE: Predicting sepsis onset with a recurrent neural network and performance comparison with InSight - a previously proposed algorithm for the prediction of sepsis onset.

Authors

  • Matthieu Scherpf
    Institute of Biomedical Engineering, TU Dresden, Germany. Electronic address: Matthieu.Scherpf@tu-dresden.de.
  • Felix Gräßer
    Institute of Biomedical Engineering, TU Dresden, Germany.
  • Hagen Malberg
    Institute of Biomedical Engineering, TU Dresden, Germany.
  • Sebastian Zaunseder
    Faculty of Information Technology, FH Dortmund, Germany.