Increased SNR efficiency in velocity selective arterial spin labeling using multiple velocity selective saturation modules (mm-VSASL).

Abstract

Velocity-selective arterial spin labeling (VSASL) is theoretically insensitive to transit delay (TD) effects. However, it uses saturation instead of inversion, resulting in compromised signal to noise ratio (SNR). In this study we explore the use of multiple velocity-selective saturation (VSS) modules in VSASL (mm-VSASL) to improve SNR. Theoretical SNR efficiency improvement and optimized parameters were calculated from simulations for mm-VSASL. VSASL with two VSS modules (VSASL-2VSS) was implemented to measure cerebral blood flow in vivo, compared with conventional VSASL (VSASL-1VSS), pulsed ASL (PASL), and pseudo-continuous ASL (PCASL). TDs and bolus durations (BDs) were measured to validate the simulations and to examine the TD sensitivity of these preparations. Compared with VSASL-1VSS, VSASL-2VSS achieved a significant improvement of SNR (22.1 ± 1.9%, P = 1.7 × 10(-6) ) in vivo, consistent with a 22.7% improvement predicted from simulations. The SNR was comparable to or higher (in gray matter, P = 4.3 × 10(-3) ) than that using PCASL. VSASL was experimentally verified to have minimal TD effects. Utilizing multiple VSS modules can improve the SNR efficiency of VSASL. Mm-VSASL may result in an SNR that is comparable to or even higher than that of PCASL in applications where long postlabeling delays are required.