Federated End-to-End Unrolled Models for Magnetic Resonance Image Reconstruction
Bioengineering · 16.03.2023 · 10 (3) · 364
DOI: 10.3390/bioengineering10030364 ↗Kurz gefasst
Fachartikel in der Zeitschrift Bioengineering, online erschienen am 16.03.2023. Verfasst von Brett R Levac und 2 weiteren Autor:innen. Beteiligte Einrichtung: Chandra Family Department of Electrical and Computer Engineering, The University of Texas at Austin, Austin, TX 78712, USA. Die Arbeit wurde bisher 17-mal zitiert (Crossref) und nennt 50 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NIH HHS, National Science Foundation, Oracle (United States).
Abstract (Originalsprache)
Image reconstruction is the process of recovering an image from raw, under-sampled signal measurements, and is a critical step in diagnostic medical imaging, such as magnetic resonance imaging (MRI). Recently, data-driven methods have led to improved image quality in MRI reconstruction using a limited number of measurements, but these methods typically rely on the existence of a large, centralized database of fully sampled scans for training. In this work, we investigate federated learning for MRI reconstruction using end-to-end unrolled deep learning models as a means of training global models across multiple clients (data sites), while keeping individual scans local. We empirically identify a low-data regime across a large number of heterogeneous scans, where a small number of training samples per client are available and non-collaborative models lead to performance drops. In this regime, we investigate the performance of adaptive federated optimization algorithms as a function of client data distribution and communication budget. Experimental results show that adaptive optimization algorithms are well suited for the federated learning of unrolled models, even in a limited-data regime (50 slices per data site), and that client-sided personalization can improve reconstruction quality for clients that did not participate in training.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Chandra Family Department of Electrical and Computer Engineering, The University of Texas at Austin, Austin, TX 78712, USA.
- Intel Corporation, Hillsboro, OR 97124, USA.
- Dell Medical School Department of Diagnostic Medicine, The University of Texas at Austin, Austin, TX 78712, USA.
- The Oden Institute for Computational Engineering and Sciences, The University of Texas at Austin, Austin, TX 78712, USA.
ORCID-Profile: Marius Arvinte, Jonathan I Tamir
Förderung
- NIH HHS
- National Science Foundation
- Oracle (United States)
- AWS
- NSF IFML
- NIH
- UT Machine Learning Lab
- AWS Machine Learning Research Award
Zitieren
APA
Levac, B. R., Arvinte, M., & Tamir, J. I. (2023). Federated End-to-End Unrolled Models for Magnetic Resonance Image Reconstruction. Bioengineering, 10(3), 364. https://doi.org/10.3390/bioengineering10030364
Vancouver
Levac BR, Arvinte M, Tamir JI. Federated End-to-End Unrolled Models for Magnetic Resonance Image Reconstruction. Bioengineering. 2023;10(3):364. doi:10.3390/bioengineering10030364
BibTeX
@article{Levac2023,
author = {Levac, Brett R and Arvinte, Marius and Tamir, Jonathan I},
title = {Federated End-to-End Unrolled Models for Magnetic Resonance Image Reconstruction},
journal = {Bioengineering},
year = {2023},
volume = {10},
number = {3},
pages = {364},
doi = {10.3390/bioengineering10030364}
}
EUVIMED