Magnetic resonance image reconstruction from undersampled measurements using a patch-based nonlocal operator

Qu, Xiaobo; Hou, Yingkun; Lam, Fan; Guo, Di; Zhong, Jianhui; Chen, Zhong

doi:10.1016/j.media.2013.09.007

Cited by 318 publications

(229 citation statements)

References 52 publications

Supporting

Mentioning

228

Contrasting

Unclassified

Order By: Relevance

“…2 , slice thickness = 5 mm). For the multi-channel brain imaging data, we first did the SENSE reconstruction with reduction factor 1 to compose the ground truth images to emulate the single-channel MRI data [36,37]. These ground truth images are with real and imaginary parts and we do the reconstruction from the retrospectively undersampled Fourier coefficients of these images.…”

Section: Numerical Experimentsmentioning

confidence: 99%

See 1 more Smart Citation

Projected Iterative Soft-Thresholding Algorithm for Tight Frames in Compressed Sensing Magnetic Resonance Imaging

Liu

Zhan

Cai

et al. 2016

IEEE Trans. Med. Imaging

Self Cite

163

View full text Add to dashboard Cite

Abstract-Compressed sensing has shown great potentials in accelerating magnetic resonance imaging. Fast image reconstruction and high image quality are two main issues faced by this new technology. It has been shown that, redundant image representations, e.g. tight frames, can significantly improve the image quality. But how to efficiently solve the reconstruction problem with these redundant representation systems is still challenging. This paper attempts to address the problem of applying iterative soft-thresholding algorithm (ISTA) to tight frames based magnetic resonance image reconstruction. By introducing the canonical dual frame to construct the orthogonal projection operator on the range of the analysis sparsity operator, we propose a projected iterative soft-thresholding algorithm (pISTA) and further accelerate it by incorporating the strategy proposed by Beck and Teboulle in 2009. We theoretically prove that pISTA converges to the minimum of a function with a balanced tight frame sparsity. Experimental results demonstrate that the proposed algorithm achieves better reconstruction than the widely used synthesis sparse model and the accelerated pISTA converges faster or comparable to the state-of-art smoothing FISTA. One major advantage of pISTA is that only one extra parameter, the step size, is introduced and the numerical solution is stable to it in terms of image reconstruction errors, thus allowing easily setting in many fast magnetic resonance imaging applications.

show abstract

Section: Numerical Experimentsmentioning

confidence: 99%

“…This criteria was previously used in MR image reconstructions [5,36]. we have an optimal reconstruction error in terms of RLNE.…”

Section: Numerical Experimentsmentioning

confidence: 99%

Projected Iterative Soft-Thresholding Algorithm for Tight Frames in Compressed Sensing Magnetic Resonance Imaging

Liu

Zhan

Cai

et al. 2016

IEEE Trans. Med. Imaging

Self Cite

163

View full text Add to dashboard Cite

show abstract

“…La segunda técnica propone resolver el problema tomando bloques de píxeles en lugar de tratarlos de forma independiente (Zhang and Blum, 1999;Li et al, 2001Li et al, , 2002Zhao et al, 2008;Aslantas and Kurban, 2010;Pagidimarry and Babu, 2011;De and Chanda, 2013;Qu et al, 2014;Bai et al, 2015;Cao et al, 2015;Calderon et al, 2016).…”

Section: Trabajos Basados En Procesamiento Por Bloques O Por Píxelesunclassified

Fusión de Imágenes Multi-Foco con Ventanas Variables

Calderón

Garnica-Carrillo

Flores

2018

Rev. iberoam. autom. inform. ind.

View full text Add to dashboard Cite

ResumenEn este artículo presentamos el Algoritmo Combinación Lineal de Imágenes con Ventanas Variables (CLI-VV) para la fusión de imágenes multi-foco. A diferencia del Algoritmo CLI-S presentado en un trabajo anterior, el algoritmo CLI-VV permite determinar automáticamente el tamañoóptimo de la ventana en cada píxel para la segmentación de las regiones con la mayor nitidez. También presentamos la generalizado el Algoritmo CLI-VV para la fusión de conjuntos de imágenes multi-foco con más de dos imágenes. A este nuevo algoritmo lo denominamos Fusión Multi-foco con Ventanas Variables (FM-VV). El Algoritmo CLI-VV se probó con 21 pares de imágenes sintéticas y 29 pares de imágenes multi-foco reales, y el Algoritmo FM-VV sobre 5 tríos de imáge-nes multi-foco. En todos los ejemplos se obtuvo un porcentaje de acierto competitivos, producidos en tiempos de ejecución menores a los reportados en la literatura. Palabras Clave: Fusión de imágenes multi-foco, Ventanas deslizantes, Imágenes integrales Multi Focus Image Fusion with variable size windows AbstractIn this paper we present the Linear Image Combination Algorithm with Variable Windows (CLI-VV) for the fusion of multifocus images. Unlike the CLI-S algorithm presented in a previous work, the CLI-VV algorithm allows to automatically determine the optimal size of the window in each pixel for the segmentation of the regions with the highest sharpness. We also present the generalized CLI-VV Algorithm for the fusion of sets of multi-focus images with more than two images. This new algorithm is called Variable Windows Multi-focus Fusion (FM-VV). The CLI-VV Algorithm was tested with 21 pairs of synthetic images and 29 pairs of real multi-focus images, and the FM-VV Algorithm on 5 trios of multi-focus images. In all the tests a competitive accuracy was obtained, with execution times lower than those reported in the literature. Keywords:Multi-focus image fusion, Sliding windows, Incremental images. IntroducciónHay muchas aplicaciones en las que se involucran imáge-nes, por ejemplo, en sistemas de seguridad, en fotografía, en apoyo al diagnóstico con imágenes médicas, en imágenes satelitales y en algunos ambientes de investigación. En ellas es deseable que las imágenes obtenidas tengan suficiente claridad o nitidez en todos los objetos de la escena. Sin embargo, lo anterior no siempre es posible pues el rango de distancias medidos desde la lente hasta los objetos que se desean capturar no es constante y las cámaras, microscopios y otros objetos que usan lentes para la captura de las imágenes, tienen limitaciones físicas que impiden que todos los objetos de la escena tengan la misma claridad en la imagen digital generada. A dicha limitación se le conoce como profundidad de campo, término que hace referencia al rango de distancias que la lente de una cáma-ra, con una apertura dada, es capaz de reproducir de manera nítida (Kuthirummal et al., 2011).En términos ideales, la luz reflejada por cada punto de la escena debería incidir sobre unúnico punto en el sensor de la cámara, sin em...

show abstract

“…Images are first divided into multiple overlapped square patches, and similar patches are searched [18,26]. Given a reference patch respectively [18].…”

Section: Shared Similarity Regions and Depth Informationmentioning

confidence: 99%

Multi-Focus Image Fusion Based on Nonsubsampled Shearlet Transform and Pulse Coupled Neural Network With Self-Similarity and Depth Information

2015

Int. J. Appl. Math. Mach. Learning

View full text Add to dashboard Cite

Combined with the shared similarity among multiple source images and depth of field in a camera, a new image fusion algorithm based on nonsubsampled shearlet transform (NSST) domain and pulse coupled neural network (PCNN) is proposed. First, NSST is utilized for decomposition of the source images, the low frequency coefficients are fused by weight votes in the structure-driven regions by using shared similarity and depth of field (SSSID), and the high frequency coefficients are fused by fired map of PCNN which motivated by larger Liu Shuaiqi et al. / IJAMML 2:1 (2015) 47-65 48 sum-modified-Laplacian (SML) based on SSSID, finally, the fusion image is gained by inverse NSST. The algorithm can both preserve the information of the source images well and suppress pixel distortion due to nonlinear operations in transform domain.Experimental results demonstrate that the proposed method outperforms the state-of-theart transform domain and PCNN fusion methods.

show abstract

Magnetic resonance image reconstruction from undersampled measurements using a patch-based nonlocal operator

Cited by 318 publications

References 52 publications

Projected Iterative Soft-Thresholding Algorithm for Tight Frames in Compressed Sensing Magnetic Resonance Imaging

Projected Iterative Soft-Thresholding Algorithm for Tight Frames in Compressed Sensing Magnetic Resonance Imaging

Fusión de Imágenes Multi-Foco con Ventanas Variables

Multi-Focus Image Fusion Based on Nonsubsampled Shearlet Transform and Pulse Coupled Neural Network With Self-Similarity and Depth Information

Contact Info

Product

Resources

About