Look-Locker Acquisition for Estimation of partial volume fractions in ASL Data

Due to the relatively low spatial resolution, arterial spin labeling (ASL) images are strongly affected by partial volume effects (PVE). This can make the qualitative analysis of ASL images difficult and it also significantly decreases the accuracy of cerebral blood flow (CBF) quantification. The PVE can be corrected for by using gray matter (GM) and white matter (WM) tissue segmentation of high-resolution T1-weighted images as proposed by Asllani et al1. The main drawback of this method is the need for precise knowledge of the partial volume ratios. However, the segmentation of T1-weighted images is very difficult at the border of GM and WM. Moreover, ASL images contain susceptibility induced deformations typical for EPI images and thus a correct registration with T1-weighted images is not always possible. Hence, the PVE correction using segmented T1-weighted images can introduce artifacts. In this study, we analyze an alternative method to obtain the partial volume ratios through the longitudinal magnetization relaxation times obtained from a Look-Locker sampling acquired in a multiple inversion time (TI) ASL sequence.