Partial <i>k</i>‐space reconstruction in single‐shot diffusion‐weighted echo‐planar imaging

Storey, Pippa; Frigo, Fred J.; Hinks, R. Scott; Mock, Bryan J.; Collick, B.; Baker, Nick; Marmurek, Jonathan; Graham, Simon J.

doi:10.1002/mrm.21132

Cited by 45 publications

(74 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If the apparent k-space center for a particular voxel is shifted beyond the edge of the acquired kspace, then the signal for that voxel will suddenly drop dramatically, in a similar manner to that previously observed due to bulk motion [Storey et al, 2008;Wedeen et al, 1994] or due to extra phase accrual due to concomitant field terms [Meier et al, 2008]. In our case the use of a (3/4) partial-Fourier acquisition means that the phase ramp need only reach p/2 per voxel in the anterior-to-posterior direction to experience this signal loss.…”

Section: Cause Of the Artifactmentioning

confidence: 68%

Addressing a systematic vibration artifact in diffusion‐weighted MRI

Gallichan

Scholz

Bartsch

et al. 2009

Human Brain Mapping

121

View full text Add to dashboard Cite

We have identified and studied a pronounced artifact in diffusion-weighted MRI on a clinical system. The artifact results from vibrations of the patient table due to low-frequency mechanical resonances of the system which are stimulated by the low-frequency gradient switching associated with the diffusion-weighting. The artifact manifests as localized signal-loss in images acquired with partial Fourier coverage when there is a strong component of the diffusion-gradient vector in the left-right direction. This signal loss is caused by local phase ramps in the image domain which shift the apparent k-space center for a particular voxel outside the covered region. The local signal loss masquerades as signal attenuation due to diffusion, severely disrupting the quantitative measures associated with diffusion-tensor imaging (DTI). We suggest a way to improve the interpretation of affected DTI data by including a co-regressor which accounts for the empirical response of regions affected by the artifact. We also demonstrate that the artifact may be avoided by acquiring full k-space data, and that subsequent increases in TE can be avoided by employing parallel acceleration.

show abstract

Section: Cause Of the Artifactmentioning

confidence: 68%

Addressing a systematic vibration artifact in diffusion‐weighted MRI

Gallichan

Scholz

Bartsch

et al. 2009

Human Brain Mapping

121

View full text Add to dashboard Cite

show abstract

“…Investigations into signal attenuations instigated by non-Brownian motion (11,(18)(19)(20)(21) have identified a number of attenuation mechanisms, each influenced by a specific set of acquisition characteristics. Consequentially, observed cardiac pulsation effects on single-shot DW-echo-planar imaging measurements range from prominent signal voids (12), to being less pronounced (9), to not being detected at all (11,13).…”

Section: Discussionmentioning

confidence: 99%

“…Subsequent phase corrections implicitly assume that phase errors can be characterized within this reduced frequency range. Violations of this assumption have been shown to lead to image artifacts (11,19). Furthermore, phase-correction algorithms generally assume that motion is that of a rigid body.…”

Section: Discussionmentioning

confidence: 99%

A quantitative analysis of the benefits of cardiac gating in practical diffusion tensor imaging of the brain

2010

View full text Add to dashboard Cite

Increasing demand for quantitative diffusion MRI in biomedical research necessitates a reevaluation of potential sources of artifact. While cardiac pulsation-induced effects have been reported previously, their quantitative characterization under standard acquisition protocols remains outstanding. To determine whether practical data quality improvements merit the use of cardiac gating, the maximum effects induced in apparent diffusion coefficient, fractional anisotropy, and mean diffusivity measurements were quantitatively estimated via measurements and bootstrapping simulations. In apparent diffusion coefficient time courses, cardiac pulsation effects were measured to occupy approximately 6% of the heartbeat duration, with considerable intersession differences in effect magnitude. Bootstrapping revealed overestimations in fractional anisotropy up to 0.26, and mean diffusivity up to 0. The increasing demand for quantitative diffusion MRI in biomedical research and its potential application in clinical practice have highlighted the need to determine and optimize the precision and accuracy of measurements.Diffusion-weighted imaging (DWI) is a motion-sensitized variant of MRI, quantifying molecular Brownian motion along a single direction. In diffusion tensor imaging (DTI), a minimum of six non-collinear DWI measurements are used to characterize three-dimensional profiles of Brownian motion. These are commonly summarized as biologically relevant quantitative diffusion tensor (DT) indices, such as fractional anisotropy (FA) and mean diffusivity (MD) (1).The inherently high motion sensitivity of DWI makes measurements potentially susceptible to contamination by non-Brownian motion. While Brownian displacements during diffusion-encoding times of 50 ms (typical for contemporary human DWI scans) are approximately 10 lm (1,2), cardiac pulsation-induced displacements of up to 130 lm to 184 lm (3,4), with velocities of up to 1.5 mm sec À1 (3) have been reported. Single-direction DWI, scalar DT indices, and tractography results measured with single-shot echo-planar imaging sequences have been observed to exhibit cardiac pulsation-induced effects (5-12). Restricting DWI acquisitions to periods of reduced cardiac pulsation-induced motion, henceforth referred to as the ''still phase'', was found to prevent the occurrence of such effects (5-10,12).In numerous investigations (5,8-10,12,13), the number of measurable repeats was boosted by minimizing the number of DWI measurements per dataset. However, the biological relevance of single-direction DWI measurements is limited, and DTs are usually based on measurements along more than the minimum six directions (14,15). Therefore, while this approach is effective at demonstrating the susceptibility of DWI and DTI measures to cardiac pulsation, and the spatial distribution of induced errors, it remains qualitative. In a pilot investigation (7), cardiac pulsation was shown to induce notable errors in DT indices measured using a standard protocol with 15 gradient directions. However, a ...

show abstract

“…Despite effectiveness in recovering spatial resolution from insuffi cient data, this method of reconstruction may introduce motion sensitivity, as the phase variation assumption is violated when patient motion is large. 6 The reduction of k-space sampling to reduce image distortion and motion sensitivity is thus a trade-off relation in half Fourier acquisition combined with phase constraint reconstruction. This motion sensitivity introduces subtle signal changes in each slice, and the signal variation causes band-like artifacts for slice direction.…”

Section: Introductionmentioning

confidence: 99%

“…To maintain the spatial resolution with reduced k-space data, phase constraint reconstruction is used. 6 Phase constraint reconstruction assumes phase variation in the image and reconstructs the image without scarifying spatial resolution from partially k-space data. Despite effectiveness in recovering spatial resolution from insuffi cient data, this method of reconstruction may introduce motion sensitivity, as the phase variation assumption is violated when patient motion is large.…”

Section: Introductionmentioning

confidence: 99%