Accelerating 3D radial MPnRAGE using a self‐supervised deep factor model
Magnetic Resonance in Medicine · 2025-09 · 94 (3) · 1191-1201
DOI: 10.1002/mrm.30549 ↗Kurz gefasst
Fachartikel in der Zeitschrift Magnetic Resonance in Medicine, online erschienen am 02.06.2025. Verfasst von Yan Chen und 6 weiteren Autor:innen. Beteiligte Einrichtung: Electrical and Computer Engineering, University of Virginia, Charlottesville, Virginia, USA. Hauptthemen laut MEDLINE: Magnetic Resonance Imaging, Deep Learning, Imaging, Three-Dimensional, Image Processing, Computer-Assisted. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 46 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NIH HHS, NICHD NIH HHS, GE Healthcare.
Abstract (Originalsprache)
Purpose
To develop a self‐supervised and memory‐efficient deep learning image reconstruction method for 4D non‐Cartesian MRI with high resolution and a large parametric dimension.
Methods
The deep factor model (DFM) represents a parametric series of 3D multicontrast images using a neural network conditioned by the inversion time using efficient zero‐filled reconstructions as input estimates. The model parameters are learned in a single‐shot learning (SSL) fashion from the k‐space data of each acquisition. A compatible transfer learning (TL) approach using previously acquired data is also developed to reduce reconstruction time. The DFM is compared to subspace methods with different regularization strategies in a series of phantom and in vivo experiments using the MPnRAGE acquisition for multicontrast imaging and quantitative estimation.
Results
DFM‐SSL improved the image quality and reduced bias and variance in quantitative estimates in both phantom and in vivo studies, outperforming all other tested methods. DFM‐TL reduced the inference time while maintaining a performance comparable to DFM‐SSL and outperforming subspace methods with multiple regularization techniques.
Conclusion
The proposed DFM offers a superior representation of the multicontrast images compared to subspace models, especially in the highly accelerated MPnRAGE setting. The self‐supervised training is ideal for methods with both high resolution and a large parametric dimension, where training neural networks can become computationally demanding without a dedicated high‐end GPU array.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Electrical and Computer Engineering, University of Virginia, Charlottesville, Virginia, USA.
- Waisman Center, University of Wisconsin-Madison, Madison, Wisconsin, USA.
- Department of Radiology, University of Iowa, Iowa City, Iowa, USA.
- Biomedical Engineering and Hawk-IDDRC, University of Iowa, Iowa City, Iowa, USA.
- Champaign Imaging LLC, Shoreview, Minnesota, USA.
- Department of Psychiatry, University of Iowa, Iowa City, Iowa, USA.
ORCID-Profile: Yan Chen, Steve R Kecskemeti, James H Holmes, Curtis A Corum, Nima Yaghoobi, Vincent A Magnotta, Mathews Jacob
Förderung
- NIH HHS
- NICHD NIH HHS
- GE Healthcare
- NIBIB NIH HHS
- NIA NIH HHS
- NIMH NIH HHS
- Eunice Kennedy Shriver National Institute of Child Health and Human Development
- National Institutes of Health
Zitieren
APA
Chen, Y., Kecskemeti, S. R., Holmes, J. H., Corum, C. A., Yaghoobi, N., Magnotta, V. A., & Jacob, M. (2025). Accelerating 3D radial MPnRAGE using a self‐supervised deep factor model. Magnetic Resonance in Medicine, 94(3), 1191-1201. https://doi.org/10.1002/mrm.30549
Vancouver
Chen Y, Kecskemeti SR, Holmes JH, Corum CA, Yaghoobi N, Magnotta VA, et al. Accelerating 3D radial MPnRAGE using a self‐supervised deep factor model. Magnetic Resonance in Medicine. 2025;94(3):1191-1201. doi:10.1002/mrm.30549
BibTeX
@article{Chen2025,
author = {Chen, Yan and Kecskemeti, Steve R and Holmes, James H and Corum, Curtis A and Yaghoobi, Nima and Magnotta, Vincent A and Jacob, Mathews},
title = {Accelerating 3D radial MPnRAGE using a self‐supervised deep factor model},
journal = {Magnetic Resonance in Medicine},
year = {2025},
volume = {94},
number = {3},
pages = {1191--1201},
doi = {10.1002/mrm.30549}
}
EUVIMED