Image space formalism of convolutional neural networks for k‐space interpolation
Magnetic Resonance in Medicine · 2025-12 · 94 (6) · 2680-2701
DOI: 10.1002/mrm.70002 ↗Kurz gefasst
Fachartikel in der Zeitschrift Magnetic Resonance in Medicine, online erschienen am 05.08.2025. Verfasst von P Dawood und 6 weiteren Autor:innen. Beteiligte Einrichtung: Experimental Physics 5, University of Würzburg, Würzburg, Germany. Hauptthemen laut MEDLINE: Neural Networks, Computer, Image Processing, Computer-Assisted, Brain, Magnetic Resonance Imaging. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 51 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
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. Theory and Methods The image space formalism for RAKI inference is employed by expressing nonlinear activations in k‐space as element‐wise multiplications with activation masks, which transform into convolutions in image space. Jacobians of the de‐aliased, coil‐combined image relative to the aliased coil images can be expressed algebraically; thus, the noise amplification is quantified analytically (g‐factor maps). We analyze the role of nonlinearity for noise resilience by controlling the degree of nonlinearity in the reconstruction model via the negative slope parameter in leaky ReLU.
Results
The analytical g‐factor maps correspond with those obtained from Monte Carlo simulations and from an auto differentiation approach for in vivo brain images. Apparent blurring and contrast loss artifacts are identified as implications of enhanced noise resilience. These residual artifacts can be traded against noise resilience by adjusting the degree of nonlinearity in the model (Tikhonov‐like regularization) in case of limited training data. The inspection of image space activations reveals an autocorrelation pattern leading to a potential center artifact.
Conclusion
The image space formalism of RAKI provides the means for analytical quantitative noise‐propagation analysis and human‐readable visualization of the effects of the nonlinear activation functions in k‐space.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Experimental Physics 5, University of Würzburg, Würzburg, Germany.
- Institute of Neuroradiology, University Hospital Erlangen, Erlangen, Germany.
- Magnetic Resonance and X-ray Imaging Department, Fraunhofer Institute for Integrated Circuits IIS, Division Development Center X-Ray Technology, Würzburg, Germany.
- Department of Internal Medicine I, University Hospital Würzburg, Würzburg, Germany.
- Molecular and Cellular Imaging, Comprehensive Heart Failure Center, University Hospital Würzburg, Würzburg, Germany.
ORCID-Profile: P Dawood, M Gram, I Homolya, P M Jakob, M Zaiss, M Blaimer
Zitieren
APA
Dawood, P., Breuer, F., Gram, M., Homolya, I., Jakob, P. M., Zaiss, M., & Blaimer, M. (2025). Image space formalism of convolutional neural networks for k‐space interpolation. Magnetic Resonance in Medicine, 94(6), 2680-2701. https://doi.org/10.1002/mrm.70002
Vancouver
Dawood P, Breuer F, Gram M, Homolya I, Jakob PM, Zaiss M, et al. Image space formalism of convolutional neural networks for k‐space interpolation. Magnetic Resonance in Medicine. 2025;94(6):2680-2701. doi:10.1002/mrm.70002
BibTeX
@article{Dawood2025,
author = {Dawood, P and Breuer, F and Gram, M and Homolya, I and Jakob, P M and Zaiss, M and Blaimer, M},
title = {Image space formalism of convolutional neural networks for k‐space interpolation},
journal = {Magnetic Resonance in Medicine},
year = {2025},
volume = {94},
number = {6},
pages = {2680--2701},
doi = {10.1002/mrm.70002}
}
EUVIMED