Inter-hospital transferability of AI: A case study on phase recognition in cholecystectomy.

Journal: Computers in biology and medicine
Published Date: May 5, 2025

Abstract

BACKGROUND: Identifying surgical phases is a crucial component of surgical workflow analysis, facilitating the automated evaluation of surgical procedures' performance and efficiency. A significant challenge in developing neural networks for surgical phase recognition lies in the scarcity of training data and the large variation in surgical techniques among surgeons. Consequently, it is imperative for these networks to possess generalization capabilities across diverse datasets. In this paper, we analyze the transferability of trained phase recognition models, using cholecystectomy as a case study.

Authors

  • Lasse Renz-Kiefel
    Fraunhofer Heinrich-Hertz-Institute, Vision & Imaging Technologies, Berlin, Germany.
  • Sebastian Lünse
    Brandenburg Medical School, Department of Surgery, University Hospital Brandenburg, Germany.
  • Rene Mantke
    Brandenburg Medical School, Department of Surgery, University Hospital Brandenburg, Germany; Faculty of Health Sciences, Joint Faculty of the Brandenburg University of Technology Cottbus - Senftenberg, the Brandenburg Medical School Theodor Fontane and the University of Potsdam & Department of Surgery, University Hospital Brandenburg, Germany.
  • Peter Eisert
    Fraunhofer Heinrich-Hertz-Institute, Vision & Imaging Technologies, Berlin, Germany; Humboldt-University Berlin, Visual Computing, Berlin, Germany.
  • Anna Hilsmann
    Fraunhofer Heinrich-Hertz-Institute, Vision & Imaging Technologies, Berlin, Germany.
  • Eric L Wisotzky
    Fraunhofer Heinrich-Hertz-Institute, Vision & Imaging Technologies, Berlin, Germany; Humboldt-University Berlin, Visual Computing, Berlin, Germany; Rostock University Medical Center, Klinik und Poliklinik für Hals-Nasen-Ohrenheilkunde, Kopf- und Halschirurgie "Otto Körner", Rostock, Germany. Electronic address: eric.wisotzky@hhi.fraunhofer.de.

Keywords

External Resources

Popular Topics
Recent Journals