Pneumothorax detection in chest radiographs: optimizing artificial intelligence system for accuracy and confounding bias reduction using in-image annotations in algorithm training.

Journal: European radiology
PMID: 33774722

Abstract

OBJECTIVES: Diagnostic accuracy of artificial intelligence (AI) pneumothorax (PTX) detection in chest radiographs (CXR) is limited by the noisy annotation quality of public training data and confounding thoracic tubes (TT). We hypothesize that in-image annotations of the dehiscent visceral pleura for algorithm training boosts algorithm's performance and suppresses confounders.

Authors

  • Johannes Rueckel
    Department of Radiology, University Hospital, LMU Munich, Munich, Germany.
  • Christian Huemmer
    X-Ray Products, Siemens Healthineers, Forchheim, Germany.
  • Andreas Fieselmann
    X-Ray Products, Siemens Healthineers, Forchheim, Germany.
  • Florin-Cristian Ghesu
  • Awais Mansoor
    Center for Infectious Disease Imaging (CIDI), Radiology and Imaging Science Department, National Institutes of Health (NIH), Bethesda, MD 20892, United States.
  • Balthasar Schachtner
    German Cancer Consortium, Heidelberg, Germany.
  • Philipp Wesp
    From the Department of Radiology, University Hospital, LMU Munich.
  • Lena Trappmann
    From the Department of Radiology, University Hospital, LMU Munich.
  • Basel Munawwar
    Department of Radiology, University Hospital, LMU Munich, Marchioninistr. 15, 81377, Munich, Germany.
  • Jens Ricke
    Department of Radiology, University Hospital Munich, Germany. Electronic address: jens.ricke@med.uni-muenchen.de.
  • Michael Ingrisch
    Department of Radiology, Ludwig-Maximilians-University Munich, Munich, Germany.
  • Bastian O Sabel
    Department of Radiology, University Hospital, LMU Munich, Munich, Germany.

Keywords

External Resources

Popular Topics
Recent Journals