Image space formalism of convolutional neural networks for k-space interpolation.

Journal: Magnetic resonance in medicine
Published Date: Aug 5, 2025

Abstract

PURPOSE: Noise resilience in image reconstructions by scan-specific robust artificial neural networks for k-space interpolation (RAKI) is linked to nonlinear activations in k-space. To gain a deeper understanding of this relationship, an image space formalism of RAKI is introduced for analyzing noise propagation analytically, identifying and characterizing image reconstruction features and to describe the role of nonlinear activations in a human-readable manner.

Authors

  • P Dawood
    Experimental Physics 5, University of Würzburg, Würzburg, Germany.
  • F Breuer
    Magnetic Resonance and X-ray Imaging Department, Fraunhofer Institute for Integrated Circuits IIS, Division Development Center X-Ray Technology, Würzburg, Germany.
  • M Gram
    Mech-Sense, Department of Gastroenterology and Hepatology, Aalborg University Hospital, Denmark.
  • I Homolya
    Molecular and Cellular Imaging, Comprehensive Heart Failure Center, University Hospital Würzburg, Würzburg, Germany.
  • P M Jakob
    Experimental Physics 5, University of Würzburg, Würzburg, Germany.
  • M Zaiss
    Magnetic Resonance Center, Max-Planck Institute for Biological Cybernetics, Tübingen, Germany.
  • M Blaimer
    Magnetic Resonance and X-ray Imaging Department, Fraunhofer Institute for Integrated Circuits IIS, Division Development Center X-Ray Technology, Würzburg, Germany.

Keywords

No keywords available for this article.

External Resources

Popular Topics
Recent Journals