Treffer: Improve robustness to mismatched sampling rate: An alternating deep low-rank approach for exponential function reconstruction and its biomedical magnetic resonance applications.

Title:
Improve robustness to mismatched sampling rate: An alternating deep low-rank approach for exponential function reconstruction and its biomedical magnetic resonance applications.
Authors:
Huang Y; Department of Electronic Science, Xiamen University-Neusoft Medical Magnetic Resonance Imaging Joint Research and Development Center, Biomedical Intelligent Cloud Research and Development Center, Fujian Provincial Key Laboratory of Plasma and Magnetic Resonance, Xiamen University, Xiamen, China., Wang Z; Department of Electronic Science, Xiamen University-Neusoft Medical Magnetic Resonance Imaging Joint Research and Development Center, Biomedical Intelligent Cloud Research and Development Center, Fujian Provincial Key Laboratory of Plasma and Magnetic Resonance, Xiamen University, Xiamen, China., Zhang X; College of Physics and Information Engineering, Fuzhou University, Fuzhou, China., Cao J; Department of Electronic Science, Xiamen University-Neusoft Medical Magnetic Resonance Imaging Joint Research and Development Center, Biomedical Intelligent Cloud Research and Development Center, Fujian Provincial Key Laboratory of Plasma and Magnetic Resonance, Xiamen University, Xiamen, China., Tu Z; Department of Electronic Science, Xiamen University-Neusoft Medical Magnetic Resonance Imaging Joint Research and Development Center, Biomedical Intelligent Cloud Research and Development Center, Fujian Provincial Key Laboratory of Plasma and Magnetic Resonance, Xiamen University, Xiamen, China., Lin M; Department of Applied Marine Physics and Engineering, College of Ocean and Earth Sciences, Xiamen University, Xiamen, China., Li L; Neusoft Medical System, China., Jiang X; Neusoft Medical System, China., Guo D; School of Computer and Information Engineering, Xiamen University of Technology, Xiamen, China., Qu X; Department of Electronic Science, Xiamen University-Neusoft Medical Magnetic Resonance Imaging Joint Research and Development Center, Biomedical Intelligent Cloud Research and Development Center, Fujian Provincial Key Laboratory of Plasma and Magnetic Resonance, Xiamen University, Xiamen, China. Electronic address: quxiaobo@xmu.edu.cn.
Source:
Journal of magnetic resonance (San Diego, Calif. : 1997) [J Magn Reson] 2025 Jul; Vol. 376, pp. 107898. Date of Electronic Publication: 2025 May 15.
Publication Type:
Journal Article
Language:
English
Journal Info:
Publisher: Elsevier Country of Publication: United States NLM ID: 9707935 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1096-0856 (Electronic) Linking ISSN: 10907807 NLM ISO Abbreviation: J Magn Reson Subsets: MEDLINE
Imprint Name(s):
Publication: 2004- : San Diego : Elsevier
Original Publication: San Diego : Academic Press, c1997-2004.
Contributed Indexing:
Keywords: Biomedical magnetic resonance; Deep learning; Exponential function; Low-rank; Optimization
Entry Date(s):
Date Created: 20250522 Date Completed: 20250606 Latest Revision: 20250606
Update Code:
20250607
DOI:
10.1016/j.jmr.2025.107898
PMID:
40403552
Database:
MEDLINE

Weitere Informationen

Undersampling accelerates signal acquisition at the expense of introducing artifacts. Removing these artifacts is a fundamental problem in signal processing and this task is also called signal reconstruction. Through modeling signals as the superimposed exponential functions, deep learning has achieved fast and high-fidelity signal reconstruction by training a mapping from the undersampled exponentials to the fully sampled ones. However, the mismatch, such as undersampling rates (25 % vs. 50 %), anatomical region (knee vs. brain), and contrast configurations (PDw vs. T <subscript>2</subscript> w), between the training and target data will heavily compromise the reconstruction. To overcome this limitation, we propose Alternating Deep Low-Rank (ADLR), which combines deep learning solvers and classic optimization solvers. Experimental validation on the reconstruction of synthetic and real-world biomedical magnetic resonance signals demonstrates that ADLR can effectively alleviate the mismatch issue and achieve lower reconstruction errors than state-of-the-art methods.
(Copyright © 2024. Published by Elsevier Inc.)

Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.