PCLR: Phase‐constrained low‐rank model for compressive diffusion‐weighted MRI

Gao, Hao; Li, Longchuan; Zhang, Kai; Zhou, Weifeng; Hu, Xiaoping

doi:10.1002/mrm.25052

Cited by 29 publications

(49 citation statements)

References 39 publications

(47 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Joint k‐q space accelerations, which undersample in both domains, have also been proposed . These methods use circulated or randomized k ‐space undersampling patterns for different q‐space locations to increase the incoherence of signal aliasing, and used CS or low rank models to recover the underlying images.…”

Section: Accelerating Diffusion Mri Acquisitionsmentioning

confidence: 99%

Image formation in diffusion MRI: A review of recent technical developments

Miller

2017

Magnetic Resonance Imaging

114

View full text Add to dashboard Cite

Diffusion magnetic resonance imaging (MRI) is a standard imaging tool in clinical neurology, and is becoming increasingly important for neuroscience studies due to its ability to depict complex neuroanatomy (eg, white matter connectivity). Single‐shot echo‐planar imaging is currently the predominant formation method for diffusion MRI, but suffers from blurring, distortion, and low spatial resolution. A number of methods have been proposed to address these limitations and improve diffusion MRI acquisition. Here, the recent technical developments for image formation in diffusion MRI are reviewed. We discuss three areas of advance in diffusion MRI: improving image fidelity, accelerating acquisition, and increasing the signal‐to‐noise ratio. Level of Evidence: 5 Technical Efficacy: Stage 1J. MAGN. RESON. IMAGING 2017;46:646–662

show abstract

Section: Accelerating Diffusion Mri Acquisitionsmentioning

confidence: 99%

Image formation in diffusion MRI: A review of recent technical developments

Miller

2017

Magnetic Resonance Imaging

114

View full text Add to dashboard Cite

show abstract

“…For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org. TBME-01027-2017-R1 2 brain DTI, which lowered the root-mean-squared error (RMSE) of diffusion-weighted images, FA and MD compared to FFT reconstruction at 4 × acceleration [27]. However, the phase correction component of this method requires dense sampling at the center of k-space, limiting the potential for further acceleration.…”

Section: Introductionmentioning

confidence: 99%

“…Huang et al combined an implicit low-rank constraint and a joint sparsity constraint to accelerate CDTI which, for real human heart data, reduced the RMSE of FA and MD at 5× acceleration compared to using basic compressed sensing, joint sparsity constraint alone and low-rank constraint alone [33]. However, the authors did not conduct evaluation on HA, nor performed a phase correction step to compensate for the drastic eddy current-induced phase inconsistency between diffusion directions that reduces correlation and weakens low-rankness (as described in [27]). …”

mentioning

confidence: 99%

“…In this work, we chose to apply a four-level symlet-4 3D wavelet transform with periodic boundary condition considering that a higher level and 3D transform both contributed to a more sparsified transform domain which enhanced the performance of compressed sensing. Note that any sparsifying transform can be used in place of in our framework.To demonstrate the strength of the proposed method over previous low-rank methods for accelerating DTI, we compared several reconstruction strategies with different phase correction methods: (I) LR Only and LR/CS constrained reconstruction with no phase correction (No PC) inspired from [33]; (II) LR Only and LR/CS constrained reconstruction with low-resolution phase correction (Low-Res PC) inspired from [27], where half of the sampled lines fully sampled the center of k-space for phase estimation and the remainder of the sampled lines were collected using lattice sampling as in [27]; and (III) LR Only and LR/CS constrained reconstruction with the proposed phase correction. We also demonstrated the advantage of using the joint constraints (LR/CS) over using only group sparsity constraint (CS Only) or only low-rank constraint (LR Only) with the proposed phase correction procedure.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Accelerated Cardiac Diffusion Tensor Imaging Using Joint Low-Rank and Sparsity Constraints

Nguyen

Christodoulou

et al. 2018

IEEE Trans. Biomed. Eng.

View full text Add to dashboard Cite

Abstract-Objective: The purpose of this manuscript is to accelerate cardiac diffusion tensor imaging (CDTI) by integrating low-rankness and compressed sensing. Methods: Diffusion-weighted images exhibit both transform sparsity and low-rankness. These properties can jointly be exploited to accelerate CDTI, especially when a phase map is applied to correct for the phase inconsistency across diffusion directions, thereby enhancing low-rankness. The proposed method is evaluated both ex vivo and in vivo, and is compared to methods using either a low-rank or sparsity constraint alone. Results: Compared to using a low-rank or sparsity constraint alone, the proposed method preserves more accurate helix angle features, the transmural continuum across the myocardium wall, and mean diffusivity at higher acceleration, while yielding significantly lower bias and higher intraclass correlation coefficient. Conclusion: Low-rankness and compressed sensing together facilitate acceleration for both ex vivo and in vivo CDTI, improving reconstruction accuracy compared to employing either constraint alone. Significance: Compared to previous methods for accelerating CDTI, the proposed method has the potential to reach higher acceleration while preserving myofiber architecture features which may allow more spatial coverage, higher spatial resolution and shorter temporal footprint in the future.Index Terms-Cardiac diffusion tensor imaging, phase correction, low-rank modeling, compressed sensing, helix angle, helix angle transmurality, mean diffusivity. I. INTRODUCTIONardiac diffusion tensor imaging (CDTI) is a powerful noninvasive tool capable of assessing the anatomic microstructure of the myocardium which is highly structured and organized into sheets of fibers making it suitable to be characterized by CDTI [1][2][3]. CDTI, performed both ex vivo [4][5][6] and in vivo [7,8], reveals a helical fiber pattern along the ventricle wall with left-handed orientation in the subepicardial region and right-handed orientation in the subendocardial region for healthy heart [9]. Such pattern can be characterized by helix angle (HA) which represents the elevated angle out of the short-axis plane, indicating the local fiber orientation. Myofibers around subendocardial regions, mid myocardium and subepicardial regions have HA> 0°, HA= 0° and HA< 0°, respectively [2]. In heart failure, the helical structure and orientation of the myocardial fibers are severely altered due to adverse remodeling [10,11].One of the major challenges for CDTI is the prolonged acquisition time, because of the multiple diffusion encoding measurements needed to robustly reconstruct the self-diffusion tensor [12]. In addition, multiple signal averages are required to maintain sufficient signal-to-noise ratio (SNR) due to signal loss caused by 2 decay and diffusion signal attenuation [13]. Though motion-induced signal loss can be effectively addressed by second-or higher-order motion compensation gradient waveforms [7,14], long acquisitions will likely incur more complex motio...

show abstract

“…More recent methods exploiting both low-rank structure and sparsity have been developed and shown to provide enhanced performance [23], [24], [29]. These methods have also been successfully applied to various MRI problems, e.g., real-time cardiac imaging [30], dynamic speech imaging [31], diffusion imaging [32], relaxometry [33] and spectroscopic imaging [34]. …”

Section: Introductionmentioning

confidence: 99%

Accelerated High-Dimensional MR Imaging With Sparse Sampling Using Low-Rank Tensors

Liu

Christodoulou

et al. 2016

IEEE Trans. Med. Imaging

129

109

View full text Add to dashboard Cite

High-dimensional MR imaging often requires long data acquisition time, thereby limiting its practical applications. This paper presents a low-rank tensor based method for accelerated high-dimensional MR imaging using sparse sampling. This method represents high-dimensional images as low-rank tensors (or partially separable functions) and uses this mathematical structure for sparse sampling of the data space and for image reconstruction from highly undersampled data. More specifically, the proposed method acquires two datasets with complementary sampling patterns, one for subspace estimation and the other for image reconstruction; image reconstruction from highly undersampled data is accomplished by fitting the measured data with a sparsity constraint on the core tensor and a group sparsity constraint on the spatial coefficients jointly using the alternating direction method of multipliers. The usefulness of the proposed method is demonstrated in MRI applications; it may also have applications beyond MRI.

show abstract

PCLR: Phase‐constrained low‐rank model for compressive diffusion‐weighted MRI

Cited by 29 publications

References 39 publications

Image formation in diffusion MRI: A review of recent technical developments

Image formation in diffusion MRI: A review of recent technical developments

Accelerated Cardiac Diffusion Tensor Imaging Using Joint Low-Rank and Sparsity Constraints

Accelerated High-Dimensional MR Imaging With Sparse Sampling Using Low-Rank Tensors

Contact Info

Product

Resources

About