Data Consistent Variational Networks for Zero-shot Self-supervised MR Reconstruction
Fürnrohr F, Wetzl J, Vornehm M, Giese D, Knoll F (2024)
Publication Type: Conference contribution
Publication year: 2024
Journal
Publisher: Springer Vieweg
Series: Informatik aktuell
City/Town: Wiesbaden
Pages Range: 316-321
Conference Proceedings Title: Bildverarbeitung für die Medizin 2024
Event location: Erlangen
ISBN: 9783658440367
DOI: 10.1007/978-3-658-44037-4_81
Abstract
Variational Networks are a common approach in deep learning-based accelerated MR reconstruction. Due to their architecture, they may however fail in enforcing data consistency.We propose an adjustment to the Variational Network, integrating an optimization block that ensures consistency with the measured kspace points. We show the superiority of the method for zero-shot self-supervised 3D reconstruction quantitatively on retrospectively undersampled knee-data, and qualitatively in prospectively undersampled MR angiography images.
Authors with CRIS profile
Involved external institutions
How to cite
APA:
Fürnrohr, F., Wetzl, J., Vornehm, M., Giese, D., & Knoll, F. (2024). Data Consistent Variational Networks for Zero-shot Self-supervised MR Reconstruction. In Andreas Maier, Thomas M. Deserno, Heinz Handels, Klaus Maier-Hein, Christoph Palm, Thomas Tolxdorff (Eds.), Bildverarbeitung für die Medizin 2024 (pp. 316-321). Erlangen, DE: Wiesbaden: Springer Vieweg.
MLA:
Fürnrohr, Florian, et al. "Data Consistent Variational Networks for Zero-shot Self-supervised MR Reconstruction." Proceedings of the Bildverarbeitung für die Medizin 2024, Erlangen Ed. Andreas Maier, Thomas M. Deserno, Heinz Handels, Klaus Maier-Hein, Christoph Palm, Thomas Tolxdorff, Wiesbaden: Springer Vieweg, 2024. 316-321.
BibTeX: Download