Reconstructing and analyzing the invariances of low-dose CT image denoising networks.

Journal: Medical physics
Published Date:

Abstract

BACKGROUND: Deep learning-based methods led to significant advancements in many areas of medical imaging, most of which are concerned with the reduction of artifacts caused by motion, scatter, or noise. However, with most neural networks being black boxes, they remain notoriously difficult to interpret, hindering their clinical implementation. In particular, it has been shown that networks exhibit invariances w.r.t. input features, that is, they learn to ignore certain information in the input data.

Authors

  • Elias Eulig
    German Cancer Research Center (DKFZ), Heidelberg, Germany.
  • Fabian Jäger
    Division of X-Ray Imaging and Computed Tomography, German Cancer Research Center (DKFZ), Heidelberg, Germany.
  • Joscha Maier
    German Cancer Research Center (DKFZ), Heidelberg, Germany.
  • Bjorn Ommer
  • Marc Kachelrieß
    German Cancer Research Center, Heidelberg, 69120, Germany.