Login Register Cart 0
Generic placeholder image

Current Medical Imaging

Editor-in-Chief

ISSN (Print): 1573-4056
ISSN (Online): 1875-6603

Back
Journal Home About Journal Editorial Board Journal Insight Current Issue Volumes/Issues
Subscribe
Research Article

Accelerating Dynamic MRI Reconstruction Using Adaptive Sequentially Truncated Higher-Order Singular Value Decomposition

Author(s): Yang Li, Qiannan Shen, Mingfeng Jiang*, Lingyan Zhu, Yongming Li, Pin Wang and Tie-Qiang Li

Volume 18, Issue 7, 2022

Published on: 01 April, 2022

Article ID: e030322201637 Pages: 12

DOI: 10.2174/1573405618666220303101900

Price: $65

Abstract

Background: Dynamic magnetic resonance imaging (dMRI) plays an important role in cardiac perfusion and functional clinical exams. However, further applications are limited by the speed of data acquisition.

Objective: A low-rank plus sparse decomposition approach is often introduced for reconstructing dynamic magnetic resonance imaging (dMRI) from highly under-sampling K-space data. In this paper, the reconstruction problem of DMR is transformed into a low-rank tensor plus sparse tensor recovery problem.

Methods: A sequentially truncated higher-order singular value decomposition method is proposed to quickly approximate the low-rank tensor space structure and learn sparse components by adding a tensor kernel norm to the low-rank tensor and a l1 norm to the sparse tensor to constrain the two parts at the same time. The optimization problem is solved by using the iterative soft-thresholding algorithm; therefore, under the premise of ensuring the accuracy of the data, the amount of computation can be effectively reduced.

Results: Compared with the state-of-the-art methods, the experimental results show that the proposed method can achieve better performance in terms of reconstruction speed and reconstruction quality on 3D and 4D dMRI datasets.

Conclusion: The multidimensional MRI time series is represented by the tensor tool and decomposed into low rank tensor terms and sparse tensor terms. The low rank spatial structure is captured by the adaptive ST-HOSVD for fast approximation and the sparse component is constrained efficiently with a sparsity transform and l1 norm. The optimization problem is solved by an iterative soft-thresholding algorithm. Through extensive 3D and 4D dMRI experiments, it is demonstrated that our method can achieve superior reconstruction performance and efficiency compared with the other three state-of-theart methods reported in the literature.

Keywords: dMRI, tensor decomposition, low-rank tensor space structure, sequentially truncated HOSVD, image reconstruction, compressed sensing.

Graphical Abstract

[1]
Otazo R, Kim D, Axel L, Sodickson DK. Combination of compressed sensing and parallel imaging for highly accelerated first-pass cardiac perfusion MRI. Magn Reson Med 2010; 64(3): 767-76.
[http://dx.doi.org/10.1002/mrm.22463] [PMID: 20535813]
[2]
Gamper U, Boesiger P, Kozerke S. Compressed sensing in dynamic MRI. Magn Reson Med 2008; 59(2): 365-73.
[http://dx.doi.org/10.1002/mrm.21477] [PMID: 18228595]
[3]
Liu Q, Wang S, Liang D. Sparse and dense hybrid representation via subspace modeling for dynamic MRI. Comput Med Imaging Graph 2017; 56: 24-37.
[http://dx.doi.org/10.1016/j.compmedimag.2017.01.007] [PMID: 28214787]
[4]
Doneva M, Mutapcic A, Lustig M. Compressed sensing (CS) workshop: Basic elements of compressed sensing. June 2011.
[5]
Qu X, Qiu T, Guo D, et al. High-fidelity spectroscopy reconstruction in accelerated NMR. Chem Commun (Camb) 2018; 54(78): 10958-61.
[http://dx.doi.org/10.1039/C8CC06132G] [PMID: 30198542]
[6]
Lustig M, Donoho D, Pauly JM. Sparse MRI: The application of compressed sensing for rapid MR imaging. Magn Reson Med 2007; 58(6): 1182-95.
[http://dx.doi.org/10.1002/mrm.21391] [PMID: 17969013]
[7]
Qu X, Hou Y, Lam F, Guo D, Zhong J, Chen Z. Magnetic resonance image reconstruction from undersampled measurements using a patch-based nonlocal operator. Med Image Anal 2014; 18(6): 843-56.
[http://dx.doi.org/10.1016/j.media.2013.09.007] [PMID: 24176973]
[8]
Peng Y, Suo J, Dai Q, Xu W. From compressed sensing to low-rank matrix recovery: Theory and applications. Acta Automatica Sinica 2013; 39(7): 981-94.
[http://dx.doi.org/10.1016/S1874-1029(13)60063-4]
[9]
Zhao B, Haldar JP, Brinegar C, Liang ZP. Low rank matrix recovery for real-time cardiac MRI. In: 2010 IEEE International Symposium on Biomedical Imaging: From Nano to Macro; 2010. Apr 14-17; Rotterdam, Netherlands; pp. 996-9.
[http://dx.doi.org/10.1109/ISBI.2010.5490156]
[10]
Lingala SG, DiBella E, Adluru G, McGann C, Jacob M. Accelerating free breathing myocardial perfusion MRI using multi coil radial k-t SLR. Phys Med Biol 2013; 58(20): 7309-27.
[http://dx.doi.org/10.1088/0031-9155/58/20/7309] [PMID: 24077063]
[11]
Fathi MF, Bakhshinejad A, Baghaie A, et al. Denoising and spatial resolution enhancement of 4D flow MRI using proper orthogonal de-composition and lasso regularization. Comput Med Imaging Graph 2018; 70: 165-72.
[http://dx.doi.org/10.1016/j.compmedimag.2018.07.003] [PMID: 30423501]
[12]
Rahmani M, Atia GK. High dimensional low rank plus sparse matrix decomposition. IEEE Trans Signal Process 2017; 65(8): 2004-19.
[http://dx.doi.org/10.1109/TSP.2017.2649482]
[13]
Otazo R, Candès E, Sodickson DK. Low-rank plus sparse matrix decomposition for accelerated dynamic MRI with separation of back-ground and dynamic components. Magn Reson Med 2015; 73(3): 1125-36.
[http://dx.doi.org/10.1002/mrm.25240] [PMID: 24760724]
[14]
Cichocki A, Mandic D, De Lathauwer L, et al. Tensor decompositions for signal processing applications: From two-way to multiway com-ponent analysis. IEEE Signal Process Mag 2015; 32(2): 145-63.
[http://dx.doi.org/10.1109/MSP.2013.2297439]
[15]
Kolda TG, Bader BW. Tensor decompositions and applications. SIAM Rev 2009; 51(3): 455-500.
[http://dx.doi.org/10.1137/07070111X]
[16]
Sidiropoulos ND, De Lathauwer L, Fu X, Huang K, Papalexakis EE, Faloutsos C. Tensor decomposition for signal processing and machine learning. IEEE Trans Signal Process 2017; 65(13): 3551-82.
[http://dx.doi.org/10.1109/TSP.2017.2690524]
[17]
Zhou P, Lu C, Lin Z, Zhang C. Tensor factorization for low-rank tensor completion. IEEE Trans Image Process 2018; 27(3): 1152-63.
[http://dx.doi.org/10.1109/TIP.2017.2762595] [PMID: 29028199]
[18]
Luo X, Zhou M, Xia Y, Zhu Q, Ammari AC, Alabdulwahab A. Generating highly accurate predictions for missing QoS data via aggregating nonnegative latent factor models. IEEE Trans Neural Netw Learn Syst 2016; 27(3): 524-37.
[http://dx.doi.org/10.1109/TNNLS.2015.2412037] [PMID: 25910255]
[19]
Wu H, Luo X, Zhou M. Advancing non-negative latent factorization of tensors with diversified regularizations. IEEE Trans Serv Comput 2020; 1.
[http://dx.doi.org/10.1109/TSC.2020.2988760]
[20]
Luo X, Wu H, Yuan H, Zhou M. Temporal pattern-aware QoS prediction via biased non-negative latent factorization of tensors. IEEE Trans Cybern 2020; 50(5): 1798-809.
[http://dx.doi.org/10.1109/TCYB.2019.2903736] [PMID: 30969935]
[21]
Gao S, Zhou M, Wang Y, Cheng J, Yachi H, Wang J. Dendritic neuron model with effective learning algorithms for classification, approxi-mation, and prediction. IEEE Trans Neural Netw Learn Syst 2019; 30(2): 601-14.
[http://dx.doi.org/10.1109/TNNLS.2018.2846646] [PMID: 30004892]
[22]
Zhang Y, Zhou P, Cui G. Multi-model based PSO method for burden distribution matrix optimization with expected burden distribution output behaviors. IEEE/CAA J Autom Sin 2018. 6(6): 1506-12.
[http://dx.doi.org/10.1109/JAS.2018.7511090]
[23]
Lv Z, Wang L, Han Z, Zhao J, Wang W. Surrogate-assisted particle swarm optimization algorithm with Pareto active learning for expensive multi-objective optimization. IEEE/CAA J Autom Sin 2019. 6(3): 838-49.
[http://dx.doi.org/10.1109/JAS.2019.1911450]
[24]
Ivanov M, Sergyienko O, Tyrsa V, et al. Influence of data clouds fusion from 3D real-time vision system on robotic group dead reckoning in unknown terrain. IEEE/CAA J Autom Sin 2020. 7(2): 368-85.
[25]
Zeng D, Xie Q, Cao W, et al. Low-dose dynamic cerebral perfusion computed tomography reconstruction via kronecker-basis-representation tensor sparsity regularization. IEEE Trans Med Imaging 2017; 36(12): 2546-56.
[http://dx.doi.org/10.1109/TMI.2017.2749212] [PMID: 28880164]
[26]
Zhong W, Li D, Wang L, Zhang M. Low-rank plus sparse reconstruction using dictionary learning for 3D-MRI. In: 2016 9th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI); 2016. Oct 15-17; Datong, China.
[27]
Liu S, Cao J, Liu H, Tan X, Zhou X. Group sparsity with orthogonal dictionary and nonconvex regularization for exact MRI reconstruction. Inf Sci 2018; 451: 161-79.
[http://dx.doi.org/10.1016/j.ins.2018.03.064]
[28]
Fu Y, Dong W. 3D magnetic resonance image denoising using low-rank tensor approximation. Neurocomputing 2016; 195: 30-9.
[http://dx.doi.org/10.1016/j.neucom.2015.09.125]
[29]
Rasch J, Kolehmainen V, Nivajärvi R, et al. Dynamic MRI reconstruction from undersampled data with an anatomical prescan. Inverse Probl 2018; 34(7): 074001.
[http://dx.doi.org/10.1088/1361-6420/aac3af]
[30]
Ulas C, Gómez PA, Sperl JI, Preibisch C, Menze BH. Spatiotemporal MRI reconstruction by enforcing local and global regularity via dynamic total variation and nuclear norm minimization. In: 2016 IEEE 13th International Symposium on Biomedical Imaging (ISBI); 2016. Apr 13-16; Prague, Czech Republic.
[http://dx.doi.org/10.1109/ISBI.2016.7493270]
[31]
Yang X, Luo Y, Chen S, Zhen X, Yu Q, Liu K. Dynamic MRI reconstruction from highly undersampled (k, t)-space data using weighted Schatten p-norm regularizer of tensor. Magn Reson Imaging 2017; 37: 260-72.
[http://dx.doi.org/10.1016/j.mri.2016.10.025] [PMID: 27832975]
[32]
Yu Y, Jin J, Liu F, Crozier S. Multidimensional compressed sensing MRI using tensor decomposition-based sparsifying transform. PLoS One 2014; 9(6): e98441.
[http://dx.doi.org/10.1371/journal.pone.0098441] [PMID: 24901331]
[33]
Roohi SF, Zonoobi D, Kassim AA, Jaremko JL. Multi-dimensional low rank plus sparse decomposition for reconstruction of under-sampled dynamic MRI. Pattern Recognit 2017; 63: 667-79.
[http://dx.doi.org/10.1016/j.patcog.2016.09.040]
[34]
Lu C, Feng J, Chen Y, Liu W, Lin Z, Yan S. Tensor robust principal component analysis with a new tensor nuclear norm. IEEE Trans Pattern Anal Mach Intell 2020; 42(4): 925-38.
[http://dx.doi.org/10.1109/TPAMI.2019.2891760] [PMID: 30629495]
[35]
Zhang C, Hu W, Jin T, Mei Z. Nonlocal image denoising via adaptive tensor nuclear norm minimization. Neural Comput Appl 2018; 29(1): 3-19.
[http://dx.doi.org/10.1007/s00521-015-2050-5]
[36]
Yokota T, Lee N, Cichocki A. Robust multilinear tensor rank estimation using higher order singular value decomposition and information criteria. IEEE Trans Signal Process 2016; 65(5): 1196-206.
[http://dx.doi.org/10.1109/TSP.2016.2620965]
[37]
Rövid A, Rudas IJ, Sergyán S, Szeidl L. HOSVD based image processing techniques. In: AIKED'11: Proceedings of the 10th WSEAS International Conference on Artificial Intelligence, Knowledge Engineering and Data Bases; 2011. Feb 20; Stevens Point, Wisconsin, United States; pp. 297-302.
[38]
Liu C, Zhou J, He K. Image compression based on truncated HOSVD. In: 2009 International Conference on Information Engineering and Computer Science; 2009. Dec 19-20; Wuhan, China; pp. 1-4.
[39]
Vannieuwenhoven N, Vandebril R, Meerbergen K. A new truncation strategy for the higher-order singular value decomposition. SIAM J Sci Comput 2012; 34(2): A1027-52.
[http://dx.doi.org/10.1137/110836067]
[40]
Balda ER, Cheema SA, Steinwandt J, Haardt M, Weiss A, Yeredor A. First-order perturbation analysis of low-rank tensor approximations based on the truncated HOSVD. In: 2016 50th Asilomar Conference on Signals, Systems and Computers; 2016. Nov 6-9; Pacific Groove, CA, USA; pp. 1723-7.
[http://dx.doi.org/10.1109/ACSSC.2016.7869677]
[41]
Fang Z, Yang X, Han L, Liu X. A sequentially truncated higher order singular value decomposition-based algorithm for tensor completion. IEEE Trans Cybern 2019; 49(5): 1956-67.
[http://dx.doi.org/10.1109/TCYB.2018.2817630] [PMID: 29993938]
[42]
Rauhut H, Schneider R, Stojanac Ž. Low rank tensor recovery via iterative hard thresholding. Linear Algebra Appl 2017; 523: 220-62.
[http://dx.doi.org/10.1016/j.laa.2017.02.028]
[43]
Zhang Y, Dong Z, Phillips P, Wang S, Ji G, Yang J. Exponential wavelet iterative shrinkage thresholding algorithm for compressed sensing magnetic resonance imaging. Inf Sci 2015; 322: 115-32.
[http://dx.doi.org/10.1016/j.ins.2015.06.017]
[44]
Wu Y, Ma Y, Liu J, Du J, Xing L. Self-attention convolutional neural network for improved MR image reconstruction. Inf Sci 2019; 490: 317-28.
[http://dx.doi.org/10.1016/j.ins.2019.03.080] [PMID: 32817993]
[45]
Wang S, Ke Z, Cheng H, et al. DIMENSION: Dynamic MR imaging with both k-space and spatial prior knowledge obtained via multi-supervised network training. NMR Biomed 2019; e4131.
[http://dx.doi.org/10.1002/nbm.4131] [PMID: 31482598]
[46]
Huang J, Song W, Wang L, Zhu Y. The influence of radial undersampling schemes on compressed sensing in cardiac DTI. Sensors (Basel) 2018; 18(7): 1-14.
[http://dx.doi.org/10.3390/s18072388] [PMID: 30041419]

Rights & Permissions Print Cite
Article Metrics
24
1
© 2024 Bentham Science Publishers | Privacy Policy