Compressed Sensing MRI using Curvelet Sparsity and Nonlocal Total Variation: CS-NLTV

Yazdanpanah, Ali Pour; Regentova, Emma E.

doi:10.2352/issn.2470-1173.2017.13.ipas-197

Cited by 8 publications

(7 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Unfortunately, ||x|| 0 in Eqs. (2) and (3) are not directly processable, because the gradient of ||x|| 0 cannot be solved, resulting in uncertainty of optimization direction and cannot be optimized. Therefore, only ergodic method can be used to solve L 0 -norm, thus greatly increasing the computational complexity.…”

Section: Arg Minmentioning

confidence: 99%

“…MR image recovery [1,2] plays an essential role in clinical diagnosis. However, at present, the quality of MR image recovery needs to be improved.…”

Section: Introductionmentioning

confidence: 99%

“…(1), we try to recover the sparse signal x from given {y, A}. In this case, the sensing matrix A contains more columns than rows, which means that there will be more than one solution that satisfies the constrain ||Ax − y|| 2 2 ≤ ( is a very small constant). This makes the recovery of sparse signal x an ill-posed problem.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A New Non-Convex Regularized Sparse Reconstruction Algorithm for Compressed Sensing Magnetic Resonance Image Recovery

Yin¹,

Wang²,

Yue³

et al. 2018

PIER C

View full text Add to dashboard Cite

Compressed sensing (CS) relies on the sparse priorin posed on the signal to solve the ill-posed recovery problem in an under-determined linear system (ULS). Motivated by the theory, this paper proposes a new algorithm called regularized re-weighted inverse trigonometric smoothed function approximating L 0-norm minimization (RRITSL0) algorithm, where the inverse trigonometric (IT) function, iteratively re-weighted scheme and regularization mechanism constitute the core of the proposed RRITSL0 algorithm. Compared with other state-of-the-art functions, our proposed IT function cluster can better approximate the L 0-norm, thus improving the reconstruction accuracy. And the new re-weighted scheme we adopted can promote sparsity and speed up convergence. Moreover, the regularization mechanism makes the RRITSL0 algorithm more robust against noise. The performance of the proposed algorithm is verified via numerical experiments with additive noise. Furthermore, the experiments prove the superiority of the RRITSL0 algorithm in magnetic resonance (MR) image recovery.

show abstract

Section: Arg Minmentioning

confidence: 99%

“…MR image recovery [1,2] plays an essential role in clinical diagnosis. However, at present, the quality of MR image recovery needs to be improved.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A New Non-Convex Regularized Sparse Reconstruction Algorithm for Compressed Sensing Magnetic Resonance Image Recovery

Yin¹,

Wang²,

Yue³

et al. 2018

PIER C

View full text Add to dashboard Cite

show abstract

“…These techniques allow for images to be reconstructed with the similar linear algebra equation that express both data similarity term and sparsity penalty term. These have allowed to increase data acquisition speed while generating better reconstructions [23,19,17,1,8,32,33,9]. Due to nondifferentiability of some regularizers, proximal methods like alternating direction method of multipliers (ADMM) [3] has been proposed.…”

Section: Introductionmentioning

confidence: 99%

Deep Plug-and-Play Prior for Parallel MRI Reconstruction

Yazdanpanah

Afacan

Warfield

2019

2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)

View full text Add to dashboard Cite

Fast data acquisition in Magnetic Resonance Imaging (MRI) is vastly in demand and scan time directly depends on the number of acquired k-space samples. Conventional MRI reconstruction methods for fast MRI acquisition mostly relied on different regularizers which represent analytical models of sparsity. However, recent data-driven methods based on deep learning has resulted in promising improvements in image reconstruction algorithms. In this paper, we propose a deep plug-and-play prior framework for parallel MRI reconstruction problems which utilize a deep neural network (DNN) as an advanced denoiser within an iterative method. This, in turn, enables rapid acquisition of MR images with improved image quality. The proposed method was compared with the reconstructions using the clinical gold standard GRAPPA method. Our results with undersampled data demonstrate that our method can deliver considerably higher quality images at high acceleration factors in comparison to clinical gold standard method for MRI reconstructions. Our proposed reconstruction enables an increase in acceleration factor, and a reduction in acquisition time while maintaining high image quality.

show abstract

“…Thus, curvelet transform can efficiently characterize anisotropic features such as curves, edges, and arcs. 46 The discrete curvelet transform was implemented using CurveLab 47 with curvelets via wrapping approach. It includes four steps: 2D fast Fourier transform (FFT) , windowing, frequency wrapping, and 2D inverse FFT.…”

mentioning

confidence: 99%

Accelerating 3D single-molecule localization microscopy using blind sparse inpainting

et al. 2021

View full text Add to dashboard Cite

Significance: Single-molecule localization-based super-resolution microscopy has enabled the imaging of microscopic objects beyond the diffraction limit. However, this technique is limited by the requirements of imaging an extremely large number of frames of biological samples to generate a super-resolution image, thus requiring a longer acquisition time. Additionally, the processing of such a large image sequence leads to longer data processing time. Therefore, accelerating image acquisition and processing in single-molecule localization microscopy (SMLM) has been of perennial interest. Aim: To accelerate three-dimensional (3D) SMLM imaging by leveraging a computational approach without compromising the resolution. Approach: We used blind sparse inpainting to reconstruct high-density 3D images from low-density ones. The low-density images are generated using much fewer frames than usually needed, thus requiring a shorter acquisition and processing time. Therefore, our technique will accelerate 3D SMLM without changing the existing standard SMLM hardware system and labeling protocol. Results: The performance of the blind sparse inpainting was evaluated on both simulation and experimental datasets. Superior reconstruction results of 3D SMLM images using up to 10-fold fewer frames in simulation and up to 50-fold fewer frames in experimental data were achieved. Conclusions: We demonstrate the feasibility of fast 3D SMLM imaging leveraging a computational approach to reduce the number of acquired frames. We anticipate our technique will enable future real-time live-cell 3D imaging to investigate complex nanoscopic biological structures and their functions.

show abstract

Compressed Sensing MRI using Curvelet Sparsity and Nonlocal Total Variation: CS-NLTV

Cited by 8 publications

References 0 publications

A New Non-Convex Regularized Sparse Reconstruction Algorithm for Compressed Sensing Magnetic Resonance Image Recovery

A New Non-Convex Regularized Sparse Reconstruction Algorithm for Compressed Sensing Magnetic Resonance Image Recovery

Deep Plug-and-Play Prior for Parallel MRI Reconstruction

Accelerating 3D single-molecule localization microscopy using blind sparse inpainting

Contact Info

Product

Resources

About