Development and evaluation of a deep learning model to improve the usability of polyp detection systems during interventions.

Journal: United European gastroenterology journal
Published Date:

Abstract

BACKGROUND: The efficiency of artificial intelligence as computer-aided detection (CADe) systems for colorectal polyps has been demonstrated in several randomized trials. However, CADe systems generate many distracting detections, especially during interventions such as polypectomies. Those distracting CADe detections are often induced by the introduction of snares or biopsy forceps as the systems have not been trained for such situations. In addition, there are a significant number of non-false but not relevant detections, since the polyp has already been previously detected. All these detections have the potential to disturb the examiner's work.

Authors

  • Markus Brand
    Interventional and Experimental Endoscopy (InExEn), Department of Internal Medicine II, University Hospital Würzburg, Würzburg, Germany.
  • Joel Troya
    Department of Internal Medicine II, Interventional and Experimental Endoscopy (InExEn), University Hospital Wuerzburg, Würzburg, Germany.
  • Adrian Krenzer
    Department of Artificial Intelligence and Knowledge Systems, University of Würzburg, Germany.
  • Zita Saßmannshausen
    Interventional and Experimental Endoscopy (InExEn), Internal Medicine II, Gastroenterology, University Hospital Würzburg, Würzburg, Germany.
  • Wolfram G Zoller
    Department of Internal Medicine and Gastroenterology, Katharinenhospital, Stuttgart, Germany.
  • Alexander Meining
    Department of Gastroenterology, University of Würzburg, Würzburg, Germany.
  • Thomas J Lux
    Interventional and Experimental Endoscopy (InExEn), Internal Medicine II, Gastroenterology, University Hospital Würzburg, Würzburg, Germany.
  • Alexander Hann
    Department of Internal Medicine II, Interventional and Experimental Endoscopy (InExEn), University Hospital Wuerzburg, Würzburg, Germany.