Anatomically constrained reconstruction from noisy data

Haldar, Justin P.; Hernando, Diego; Song, Sheng; Liang, Zhi Pei

doi:10.1002/mrm.21536

Cited by 104 publications

(130 citation statements)

References 40 publications

Supporting

Mentioning

124

Contrasting

Order By: Relevance

“…Many choices can be made for Ψ 1,2 (·) to incorporate prior information about the unknown spatiospectral/spatiotemporal function (e.g., those in (19,23)). In this work, we choose the following regularization form for the metabolite signal component [8] where D is a finite difference operator, W contains edge weights derived from highresolution anatomical images (17) and Ψ denotes a temporal sparsifying transform (e.g., the Fourier transform for MRSI (24)). This choice is motivated by the advantages of such edgepreserving (or non-quadratic) penalties shown in recent developments for sparse sampling and denoising.…”

Section: Image Reconstructionmentioning

confidence: 99%

“…Significant efforts have been made in the last several decades to achieve fast, high-resolution MR spectroscopic imaging (MRSI), through the development of fast sequences (1)(2)(3)(4)(5)(6)(7)(8)(9)(10) and advanced image reconstruction methods (11)(12)(13)(14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25). SPICE (SPectroscopic Imaging by exploiting spatiospectral CorrElation) is a relatively new approach that we recently proposed to achieve high-resolution MRSI with good signal-to-noise ratio (SNR) and speed (26).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

High‐resolution ¹H‐MRSI of the brain using SPICE: Data acquisition and image reconstruction

Lam

Clifford

et al. 2016

Magnetic Resonance in Med

Self Cite

View full text Add to dashboard Cite

Purpose-To develop data acquisition and image reconstruction methods to enable highresolution 1 H MR spectroscopic imaging (MRSI) of the brain, using the recently proposed subspace-based spectroscopic imaging framework called SPICE (SPectroscopic Imaging by exploiting spatiospectral CorrElation).Theory and Methods-SPICE is characterized by the use of a subspace model for both data acquisition and image reconstruction. For data acquisition, we propose a novel spatiospectral encoding scheme that provides hybrid data sets for determining the subspace structure and for image reconstruction using the subspace model. More specifically, we use a hybrid chemical shift imaging (CSI)/echo-planar spectroscopic imaging (EPSI) sequence for two-dimensional (2D) MRSI and a dual-density, dual-speed EPSI sequence for three-dimensional (3D) MRSI. For image reconstruction, we propose a method that can determine the subspace structure and the highresolution spatiospectral reconstruction from the hybrid data sets generated by the proposed sequences, incorporating field inhomogeneity correction and edge-preserving regularization.Results-Phantom and in vivo brain experiments were performed to evaluate the performance of the proposed method. For 2D MRSI experiments, SPICE is able to produce high SNR spatiospectral distributions with an approximately 3 mm nominal in-plane resolution from a 10-min acquisition. For 3D MRSI experiments, SPICE is able to achieve an approximately 3 mm inplane and 4 mm through-plane resolution in about 25 min.Conclusion-Special data acquisition and reconstruction methods have been developed for highresolution 1 H-MRSI of the brain using SPICE. Using these methods, SPICE is able to produce spatiospectral distributions of 1 H metabolites in the brain with high spatial resolution, while maintaining a good SNR. These capabilities should prove useful for practical applications of SPICE.

show abstract

Section: Image Reconstructionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

High‐resolution ¹H‐MRSI of the brain using SPICE: Data acquisition and image reconstruction

Lam

Clifford

et al. 2016

Magnetic Resonance in Med

Self Cite

View full text Add to dashboard Cite

show abstract

“…For example, iterative Krylov subspace solvers like the conjugate gradient (CG) method can be used. Stone et al (2008) made a GPU implementation of a CG algorithm to enable practical use of the anatomically constrained reconstruction algorithm presented by Haldar et al (2008). Two years later, this work was extended to include correction for field inhomogeneities (Zhuo et al, 2010b), multi-GPU support (Zhuo et al, 2010c) and a regularization term for spatial smoothness (Zhuo et al, 2010a).…”

Section: Mrimentioning

confidence: 99%

Medical image processing on the GPU – Past, present and future

Eklund

Dufort

Forsberg

et al. 2013

Medical Image Analysis

349

198

View full text Add to dashboard Cite

Graphics processing units (GPUs) are used today in a wide range of applications, mainly because they can dramatically accelerate parallel computing, are affordable and energy efficient. In the field of medical imaging, GPUs are in some cases crucial for enabling practical use of computationally demanding algorithms. This review presents the past and present work on GPU accelerated medical image processing, and is meant to serve as an overview and introduction to existing GPU implementations. The review covers GPU acceleration of basic image processing operations (filtering, interpolation, histogram estimation and distance transforms), the most commonly used algorithms in medical imaging (image registration, image segmentation and image denoising) and algorithms that are specific to individual modalities (CT, PET, SPECT, MRI, fMRI, DTI, ultrasound, optical imaging and microscopy). The review ends by highlighting some future possibilities and challenges.

show abstract

“…19 F imaging with MRI offers the possibility to incorporate anatomical knowledge of the problem derived from standard proton images acquired concurrently with fluorine images. For examples from closely related fields, Haldar et al [50] incorporated anatomical information into constrained reconstructions of low-SNR MR diffusion data, and Dewaraja et al [51] showed that incorporating boundary information from CT images into iterative SPECT reconstructions can improve the resultant images.…”

Section: Model-based Medical Imaging Reconstructionmentioning

confidence: 99%

“…anatomical data provided by CT [51,[83][84][85]. Furthermore, there are MRI analogies to the class of super-resolution-through-image-denoising methods [50] and to techniques that have been used to jointly estimate images obtained with differing contrasts [86]. Since there should be a high correlation between where we observe fluorine signal and organ and tissue boundaries in the body, many of these ideas are directly applicable to the secondary nucleus situation.…”

Section: Introductionmentioning

confidence: 99%

New Cardiovascular Applications of the Turbo Spin Echo Sequence in Magnetic Resonance Imaging

Fielden¹

View full text Add to dashboard Cite

The advent of fast sequences based on the Rapid Acquisition with Relaxation Enhancement (RARE) sequence allowed T 2 -weighted imaging to be performed in clinically feasible scan times and these sequences have become the workhorse for spin-echo imaging in Magnetic Resonance Imaging (MRI). In this work, we propose several variants and applications of the 3D Turbo Spin Echo (TSE) sequence designed to efficiently image endogenous and exogenous long-T 2 species in the body.There has been renewed interest in developing and utilizing non-contrast MRA methods because of the risk of nephrogenic systemic fibrosis (NSF) in patients with kidney dysfunction. TSE sequences are capable of generating excellent contrast between blood and surrounding tissue based on T 2 differences, but conventional TSE sequences are susceptible to flow-related artifacts and signal loss.Here, a new, hybrid pulse sequence that shares characteristics with the TSE and balanced Steady State Free Precession (bSSFP) sequence is introduced. Desirable T 2 contrast is provided by the TSE-like pulse sequence design and the improved flow-performance characteristics of bSSFP imaging are gained through fully-refocused gradients and refocusing ii iii pulse phase alternation of RF pulses. The principles of flow-independent angiography in the context of this new TSE sequence, including a description of tissue contrast properties, are described. Additionally, a method for fat/water separation based on phase detection is presented for this application which relies on catalyzation of the TSE echo train. This sequence is capable of performing non-contrast enhanced angiography of the peripheral arterial tree by exploiting the T 2 differences between blood and surrounding tissue.In order to take advantage of some of the unique properties afforded by non-Cartesian imaging, a 3D TSE sequence which utilizes spiral readouts was developed. For peripheral angiography, spoiler gradients placed in the through-plane direction remove spurious fat signal, and longer echo spacings are available due to spiral sampling, leading to improved venous suppression. To most efficiently acquire the data, partial k-space coverage along with variable density spirals are used to undersample k-space in a controlled and well-understood manner. Additional techniques to prevent spiral artifacts (blurring) from concomitant and B 0 fields as well as corrections to spiral trajectories are applied.The major focus of MRI research has been on improving resolution, contrast, and shortening acquisition times of standard anatomical images; however, patient outcomes may also be improved by acquiring secondary, functional information about the patient. Yet functionalizing MRI is still in early stages, and much work needs to be done before it is ready to enter the clinic in an impactful way. In a departure from angiography, the TSE sequence is used as an efficient method to generate and utilize signal from fluorine, a secondary nucleus which is of interest for cellular and molecular imaging. Here, th...

show abstract

Anatomically constrained reconstruction from noisy data

Cited by 104 publications

References 40 publications

High‐resolution ¹H‐MRSI of the brain using SPICE: Data acquisition and image reconstruction

High‐resolution ¹H‐MRSI of the brain using SPICE: Data acquisition and image reconstruction

Medical image processing on the GPU – Past, present and future

New Cardiovascular Applications of the Turbo Spin Echo Sequence in Magnetic Resonance Imaging

Contact Info

Product

Resources

About

Anatomically constrained reconstruction from noisy data

Cited by 104 publications

References 40 publications

High‐resolution 1H‐MRSI of the brain using SPICE: Data acquisition and image reconstruction

High‐resolution 1H‐MRSI of the brain using SPICE: Data acquisition and image reconstruction

Medical image processing on the GPU – Past, present and future

New Cardiovascular Applications of the Turbo Spin Echo Sequence in Magnetic Resonance Imaging

Contact Info

Product

Resources

About

High‐resolution ¹H‐MRSI of the brain using SPICE: Data acquisition and image reconstruction

High‐resolution ¹H‐MRSI of the brain using SPICE: Data acquisition and image reconstruction