Improved PASL EPI acquisitions with parallel imaging and unfold

Abstract

Pulsed Arterial Spin Labeling (PASL) has been shown to be an effective method for quantitatively measuring cerebral blood flow. Quantification errors with PASL may result from patient motion, however, so Echo Planar Imaging (EPI) is commonly used to acquire PASL images due to its high temporal resolution. EPI achieves high temporal resolution through long echo trains, which can result in image distortions due to magnetic field inhomogeneities. Thus, perfusion weighted EPI could be further improved with shorter echo trains and faster image acquisitions. This paper investigates the use of UNFOLD and parallel MR imaging to reduce EPI image acquisition time. We demonstrate how the use of varied sampling patterns and UNFOLD can completely remove the 'Nyquist ghosts' that are common in EPI acquisitions. This allows one to greatly improve the effectiveness of parallel imaging in EPI, and to reduce field inhomogeneity artifacts by reducing the echo train length.

Multiresolution MRI temperature monitoring in a reduced field of view

“…The images of the EPI sequence were corrected using the PLACE method combined with UNFOLD (30) to remove Nyquist ghosting artifacts. In the PLACE (31) method, the polarity of the sampling trajectory is alternated from one frame to the next and k-space lines in each frame are separated into negative and positive lines to form two set of images.…”

A method for continuous accelerated echo-planar imaging with self-referenced parallel MR reconstruction and artifact correction

“…Furthermore, this approach is so effective at Nyquist ghost removal, that immediate benefits to self-referenced pMRI reconstructions of accelerated EPI data can be observed [7]. Image reconstruction quality in pMRI is dependent on highquality calibration data [9].…”

“…In [7], we proposed the use of UNFOLD [8] with the EPI trajectory shifting of the PLACE method to provide Nyquist ghost removal while maintaining nearly all of the original temporal bandwidth. As described in [5], when the images associated with positive and negative readout gradients are phase aligned and added together, any residual Nyquist ghosts effectively cancel.…”