Leveraging machine learning for duration of surgery prediction in knee and hip arthroplasty - a development and validation study.

Journal: BMC medical informatics and decision making
PMID:

Abstract

BACKGROUND: Duration of surgery (DOS) varies substantially for patients with hip and knee arthroplasty (HA/KA) and is a major risk factor for adverse events. We therefore aimed (1) to identify whether machine learning can predict DOS in HA/KA patients using retrospective data available before surgery with reasonable performance, (2) to compare whether machine learning is able to outperform multivariable regression in predictive performance and (3) to identify the most important predictor variables for DOS both in a multi- and single-hospital context.

Authors

  • Benedikt Langenberger
    Department of Health Care Management, Technische Universität Berlin, Berlin, Germany.
  • Daniel Schrednitzki
    Department of Orthopaedic, Trauma, Hand and Reconstructive Surgery, Sana Klinikum Lichtenberg, Berlin, Germany.
  • Andreas Halder
    Department of Orthopedic Surgery, Sana Klinken Sommerfeld, Brandenburg, Germany.
  • Reinhard Busse
    Department of Healthcare Management, Technische Universität Berlin, 10623, Berlin, Germany.
  • Christoph Pross
    Department of Health Care Management, Technische Universität Berlin, Berlin, Germany.