This paper reports the development of a Python Non-Uniform Fast Fourier Transform (PyNUFFT) package, which accelerates non-Cartesian image reconstruction on heterogeneous platforms. Scientific computing with Python encompasses a mature and integrated environment. The NUFFT algorithm has been extensively used for non-Cartesian image reconstruction but previously there was no native Python NUFFT library. The current PyNUFFT software enables multi-dimensional NUFFT on heterogeneous platforms. The PyNUFFT also provides several solvers, including the conjugate gradient method, ℓ1 total-variation regularized ordinary least square (L1TV-OLS) and ℓ1 total-variation regularized least absolute deviation (L1TV-LAD). Metaprogramming libraries were employed to accelerate PyNUFFT. The PyNUFFT package has been tested on multi-core CPU and GPU, with acceleration factors of 6.3 -9.5× on a 32 thread CPU platform and 5.4 -13× on the GPU. Highlights• A Python non-uniform fast Fourier transform (PyNUFFT) package was developed.• Computations are accelerated on heterogeneous systems, including multicore CPU and GPU.• It provides the accelerated nonlinear conjugate gradient method, ℓ1 total-variation regularized ordinary least square and ℓ1 total-variation regularized least absolute deviation.• Pre-indexing enables multi-dimensional NUFFT and fast image gradients on heterogeneous platforms.• The single-precision version is dual-licensed under MIT and LGPL-3.0.• Applications to magnetic resonance imaging reconstruction are presented.
Intrascan subject movement in clinical MR spectroscopic examinations may result in inconsistent water suppression that distorts the metabolite signals, frame-to-frame variations in spectral phase and frequency, and consequent reductions in the signal-to-noise ratio due to destructive averaging. Frameto-frame phase/frequency corrections, although reported to be successful in achieving constructive averaging, rely on consistent water suppression, which may be difficult in the presence of intrascan motion. In this study, motion correction using nonwater-suppressed data acquisition is proposed to overcome the above difficulties. The time-domain matrix-pencil postprocessing method was used to extract water signals from the non-water-suppressed spectroscopic data, followed by phase and frequency Inevitable subject motions during successive data acquisition in magnetic resonance spectroscopy (MRS) may cause signal loss because of two major mechanisms (1-4). First, motion-induced magnetic field drifts may lead to frequency shift over multiple MRS acquisition frames. Second, subject movements in the presence of the spoiler gradients may result in inconsistent phase variations among acquisition frames. In conventional MRS studies, data from all acquisition frames are simply averaged, and therefore the destructively averaged spectrum would be affected by signal-to-noise ratio (SNR) attenuation and spectral shape distortion because of frame-to-frame phase/ frequency variations. It has been shown that, by performing frame-by-frame corrections for the intrascan phase/ frequency variations, the constructively averaged MRS data exhibit significantly improved SNR and spectral quality (1-8). The search for an effective intrascan phase/frequency correction method is thus an active field of research for in vivo MRS.Over the past decades, several methods have been proposed to measure the intrascan phase/frequency variations in MRS scans in order to perform constructive averaging (1,2,4-6). In one of these methods, the non-water-suppressed (NWS) free induction decay and water-suppressed (WS) MRS data were acquired in an interleaved manner (2). Intrascan phase/frequency variations estimated from the NWS free induction decay were used to correct the WS MRS data. In this approach, however, the phase variations in WS MRS data are corrected by values estimated at different time points. Therefore, the effectiveness of the phase correction might be suboptimal, particularly in the presence of abrupt intrascan movements.Alternatively, the phase variations induced by intrascan motion could be measured on a frame-by-frame basis directly from the residual water signals of WS MRS data (1,4). However, even though the information estimated directly from the WS MRS data better reflects the instant phase variation in comparison to the interleaved free induction decay-based measurements, this approach has two main limitations: First, the residual water signals in WS MRS data have low SNR that limits the accuracy of frameby-frame phase estimation (1). ...
Carotid artery atherosclerosis is an important source of mortality and morbidity in the Western world with significant socioeconomic implications. The quest for the early identification of the vulnerable carotid plaque is already in its third decade and traditional measures, such as the sonographic degree of stenosis, are not selective enough to distinguish those who would really benefit from a carotid endarterectomy. MRI of the carotid plaque enables the visualization of plaque composition and specific plaque components that have been linked to a higher risk of subsequent embolic events. Blood suppressed T 1 and T 2 weighted and proton density-weighted fast spin echo, gradient echo and time-of-flight sequences are typically used to quantify plaque components such as lipid-rich necrotic core, intraplaque haemorrhage, calcification and surface defects including erosion, disruption and ulceration. The purpose of this article is to review the most important recent advances in MRI technology to enable better diagnostic carotid imaging.Internal carotid artery atherosclerosis is a significant source of mortality and morbidity in the Western world.1,2 MRI enables the visualization of plaque composition, and specific plaque constituents have been linked to a higher risk of subsequent embolic events. Currently, either the sonographic or angiographic degree of carotid stenosis is used as a marker of severity for assessing carotid disease and risk of stroke. 3,4 However, there is mounting evidence that suggests that the degree of stenosis is not enough to accurately characterize carotid plaque burden and vulnerability.High-resolution, multicontrast carotid MRI protocols have been used to depict atherosclerotic components within carotid plaques, the validity of these protocols have been evaluated using histopathology, and the protocols have been applied in multicentre trials.5-7 Blood suppressed T 1 and T 2 weighted and proton density (PD)-weighted fast spin echo (FSE) and gradient echo time-of-flight sequences are typically used 8 to quantify plaque components such as lipidrich necrotic core (LRNC), 5,9-11 intraplaque haemorrhage (IPH), 5,12-15 calcification 5,9 and surface defects including erosions, disruption and ulceration. 11,[16][17][18] In addition, the intrareader, interreader 14,[19][20][21] and the interscan reproducibility 20,22,23 of quantitative measures associated with both morphology and composition have been reported. Novel MR-defined plaque features of vulnerability are emerging and appear promising for the identification of the vulnerable plaque. A recent systematic review of 9 studies with 779 subjects demonstrated that the presence of IPH, LRNC and thinning/rupture of the fibrous cap (FC) is linked to an increased risk of future stroke or transient ischaemic attack (TIA). 24 The hazard ratios for each of them as predictors of subsequent ischaemic events were 4.59 [95% confidence interval (CI), 2.91-7.24], 3.00 (95% CI, 1.51-5.95) and 5.93 (95% CI, 2.65-13.20), respectively. In a separate meta-a...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.