Early detection of sepsis utilizing deep learning on electronic health record event sequences.

Journal: Artificial intelligence in medicine
Published Date:

Abstract

BACKGROUND: The timeliness of detection of a sepsis incidence in progress is a crucial factor in the outcome for the patient. Machine learning models built from data in electronic health records can be used as an effective tool for improving this timeliness, but so far, the potential for clinical implementations has been largely limited to studies in intensive care units. This study will employ a richer data set that will expand the applicability of these models beyond intensive care units. Furthermore, we will circumvent several important limitations that have been found in the literature: (1) Model evaluations neglect the clinical consequences of a decision to start, or not start, an intervention for sepsis. (2) Models are evaluated shortly before sepsis onset without considering interventions already initiated. (3) Machine learning models are built on a restricted set of clinical parameters, which are not necessarily measured in all departments. (4) Model performance is limited by current knowledge of sepsis, as feature interactions and time dependencies are hard-coded into the model.

Authors

  • Simon Meyer Lauritsen
    Enversion A/S, Fiskerivej 12, 8000 Aarhus C, Denmark; Department of Clinical Medicine, Aarhus University, Denmark. Electronic address: sla@enversion.dk.
  • Mads Ellersgaard Kalør
    Enversion A/S, Fiskerivej 12, 8000 Aarhus C, Denmark.
  • Emil Lund Kongsgaard
    Enversion A/S, Fiskerivej 12, 8000 Aarhus C, Denmark.
  • Katrine Meyer Lauritsen
    Department of Clinical Medicine, Aarhus University, Denmark; Department of Endocrinology and Internal Medicine, Aarhus University Hospital, Aarhus, Denmark.
  • Marianne Johansson Jørgensen
    Department of Research, Horsens Regional Hospital, Denmark.
  • Jeppe Lange
    Department of Clinical Medicine, Aarhus University, Denmark; Department of Research, Horsens Regional Hospital, Denmark.
  • Bo Thiesson
    Enversion A/S, Fiskerivej 12, 8000 Aarhus C, Denmark; Department of Engineering, Aarhus University School of Engineering, Denmark.