Accelerating multi-coil MR image reconstruction using weak supervision

Deep-learning-based MR image reconstruction in settings where large fully sampled dataset collection is infeasible requires methods that effectively use both under-sampled and fully sampled datasets. This paper evaluates a weakly supervised, multi-coil, physics-guided approach to MR image reconstruc...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Magma (New York, N.Y.) Jg. 38; H. 1; S. 37 - 51
Hauptverfasser: Atalık, Arda, Chopra, Sumit, Sodickson, Daniel K.
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Cham Springer International Publishing 01.02.2025
Schlagworte:
Basic Science - Reconstruction algorithms and artificial intelligence
Biomedical Engineering and Bioengineering
Computer Appl. in Life Sciences
Deep Learning
Health Informatics
Humans
Image Processing, Computer-Assisted - methods
Imaging
Magnetic Resonance Imaging
Medicine
Medicine & Public Health
Radiology
Research Article
Solid State Physics
Supervised Machine Learning
Accelerated imaging
Weak supervision
Transfer learning
MR image reconstruction
Machine learning
Self-supervised learning
ISSN:1352-8661, 1352-8661
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Deep-learning-based MR image reconstruction in settings where large fully sampled dataset collection is infeasible requires methods that effectively use both under-sampled and fully sampled datasets. This paper evaluates a weakly supervised, multi-coil, physics-guided approach to MR image reconstruction, leveraging both dataset types, to improve both the quality and robustness of reconstruction. A physics-guided end-to-end variational network (VarNet) is pretrained in a self-supervised manner using a 4 × under-sampled dataset following the self-supervised learning via data undersampling (SSDU) methodology. The pre-trained weights are transferred to another VarNet, which is fine-tuned using a smaller, fully sampled dataset by optimizing multi-scale structural similarity (MS-SSIM) loss in image space. The proposed methodology is compared with fully self-supervised and fully supervised training. Reconstruction quality improvements in SSIM, PSNR, and NRMSE when abundant training data is available (the high-data regime), and enhanced robustness when training data is scarce (the low-data regime) are demonstrated using weak supervision for knee and brain MR image reconstructions at 8 × and 10 × acceleration, respectively. Multi-coil physics-guided MR image reconstruction using both under-sampled and fully sampled datasets is achievable with transfer learning and fine-tuning. This methodology can provide improved reconstruction quality in the high-data regime and improved robustness in the low-data regime at high acceleration rates.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:1352-8661
1352-8661
DOI:10.1007/s10334-024-01206-2