Self-supervised learning for improved calibrationless radial MRI with NLINV-Net.

Journal: Magnetic resonance in medicine
PMID:

Abstract

PURPOSE: To develop a neural network architecture for improved calibrationless reconstruction of radial data when no ground truth is available for training.

Authors

  • Moritz Blumenthal
    Institute for Diagnostic and Interventional Radiology, University Medical Center Göttingen, Göttingen, Germany.
  • Chiara Fantinato
    Institute of Biomedical Imaging, Graz University of Technology, Graz, Austria.
  • Christina Unterberg-Buchwald
    Institute for Diagnostic and Interventional Radiology, Universitätsmedizin Göttingen, Göttingen, Germany.
  • Markus Haltmeier
    Department of Mathematics, University of Innsbruck, Innsbruck, 6020, Austria.
  • Xiaoqing Wang
    Department of Anesthesiology and Operation, The First Hospital of Lanzhou University, Lanzhou, Gansu, China.
  • Martin Uecker
    Institute for Diagnostic and Interventional Radiology, University Medical Center Göttingen, Göttingen, Germany.