Accelerated MR diffusion tensor imaging using distributed compressed sensing

Wu, Yin; Zhu, Yanjie; Tang, Qiu-Yang; Zou, Chao; Liu, Wei; Dai, Ruibin; Liu, Xin; Wu, Ed X.; Ying, Leslie; Liang, Dong

doi:10.1002/mrm.24721.

Cited by 49 publications

(59 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To evaluate the performance of the proposed method, a comparison was performed with several stateof-the-art reconstruction methods, including the basic CS method [12], low rank [15], and joint sparsity method [9]. The root mean square errors (RMSE) of fractional anisotropy (FA) and mean diffusivity (MD) were calculated for each method.…”

Section: Resultsmentioning

confidence: 99%

“…When the different images are stacked as column vectors of matrix X, the resulting matrix is row sparse. Exploiting the joint sparsity property along diffusion directions can achieve better reconstruction accuracy [9].…”

Section: Theory and Algorithmmentioning

confidence: 99%

“…Considering that the diffusion weighted (DW) images obtained at different gradient directions are often correlated, a distributed compressed sensing-based method offers an alternative to reconstructing images one-by-one [9]. This technique exploits joint sparsity among DW images to obtain better reconstruction quality.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Cardiac diffusion tensor imaging based on compressed sensing using joint sparsity and low-rank approximation

Huang

Wang

Chu

et al. 2016

THC

View full text Add to dashboard Cite

Abstract. Diffusion tensor magnetic resonance (DTMR) imaging and diffusion tensor imaging (DTI) have been widely used to probe noninvasively biological tissue structures. However, DTI suffers from long acquisition times, which limit its practical and clinical applications. This paper proposes a new Compressed Sensing (CS) reconstruction method that employs joint sparsity and rank deficiency to reconstruct cardiac DTMR images from undersampled k-space data. Diffusion-weighted images acquired in different diffusion directions were firstly stacked as columns to form the matrix. The matrix was row sparse in the transform domain and had a low rank. These two properties were then incorporated into the CS reconstruction framework. The underlying constrained optimization problem was finally solved by the first-order fast method. Experiments were carried out on both simulation and real human cardiac DTMR images. The results demonstrated that the proposed approach had lower reconstruction errors for DTI indices, including fractional anisotropy (FA) and mean diffusivities (MD), compared to the existing CS-DTMR image reconstruction techniques.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Theory and Algorithmmentioning

confidence: 99%

See 1 more Smart Citation

Cardiac diffusion tensor imaging based on compressed sensing using joint sparsity and low-rank approximation

Huang

Wang

Chu

et al. 2016

THC

View full text Add to dashboard Cite

show abstract

“…This works by exploiting signal sparsity in a transform domain in which undersampling artifacts are incoherent [17,18]. Compressed sensing approaches have a wide range of applications [19][20][21] and have been used in CDTI to provide precise measurements of fractional anisotropy (FA), mean diffusivity (MD), and the primary eigenvector (Δ ) until 4× acceleration [22].Another approach is low-rank modeling (LR) which exploits signal correlation using partial separability model [23,24]. It has previously been used for diffusion-weighed image denoising in the brain [25] This is the author's version of an article that has been published in this journal.…”

mentioning

confidence: 99%

“…Low-rankness can be restored by modeling these phase inconsistencies in the form of a unit-magnitude phase map ∈ {ℂ × : | | = 1, ∀ , } which in previous work has been calculated from low-resolution scans [27]. The phase-corrected image model is thus = ∘ ( ), where ∘ denotes Hadamard (elementwise) multiplication.2) Group sparsity constraint: Group sparsity modeling is inspired by distributed compressed sensing and has been applied on its own to accelerate CDTI [22]. The underlying assumption is that an individual image in the diffusion-weighted image series not only has a sparse property in some transform domain calculated by applying the matrix , but also shares similar sparse support with other images.…”

mentioning

confidence: 99%

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

Accelerating cardiac diffusion tensor imaging using Swin transformer with octave convolution

Li,

Mo,

Wang

et al. 2024

Int J Imaging Syst Tech

View full text Add to dashboard Cite

Cardiac diffusion tensor imaging (cDTI) is a promising and noninvasive magnetic resonance imaging (MRI) technique for assessing myocardial fiber microstructure. However, its clinical application is hampered by lengthy scanning and susceptibility to motion artifacts. This study proposes a cDTI reconstruction model (OSDTI), which leverages octave convolution (OctConv) and Swin Transformer to reconstruct diffusion‐weighted (DW) images from undersampled k‐space data. Our model capitalizes on the strengths of both OctConv and Swin Transformers. Specifically, the OctConv module excels at capturing intricate details and texture information within DW images, while the residual‐connected Swin Transformer module facilitates information retention across various feature extraction stages, thereby enhancing overall image comprehension and reconstruction accuracy. The evaluation of OSDTI using a human ex vivo heart dataset robustly validated the generalization ability of the model. Comparative analyze against existing deep‐learning methods demonstrated OSDTI's superior performance in both DW image quality and diffusion tensor metrics.

show abstract

Accelerated MR diffusion tensor imaging using distributed compressed sensing

Abstract: Distributed compressed sensing is shown to be able to accelerate DTI and may be used to reduce DTI acquisition time practically.

Cited by 49 publications

References 51 publications

Cardiac diffusion tensor imaging based on compressed sensing using joint sparsity and low-rank approximation

Cardiac diffusion tensor imaging based on compressed sensing using joint sparsity and low-rank approximation

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

Accelerating cardiac diffusion tensor imaging using Swin transformer with octave convolution

Contact Info

Product

Resources

About