Applications and limitations of whole‐brain MAGIC VASO functional imaging

Open Access
Publisher Website
Google Scholar
Abstract
This work extends the multiple acquisitions with global inversion cycling vascular space occupancy (MAGIC VASO) method to human whole‐brain functional magnetic resonance imaging (fMRI) at 3.0 Tesla and demonstrates the need to consider the dynamic contribution of cerebrospinal fluid (CSF) to the relative VASO signal change (ΔVASO/VASO). Simulations were performed to determine the optimal slice number between global inversions, and correction factors were obtained to account for incomplete blood nulling in particular slices. The necessity of an accurate estimate of resting cerebral blood volume (CBVrest) is discussed in the context of ΔCBV/CBV calculations. A three‐compartment model is proposed to include both the resting and changing fractional CSF contribution (xc,rest and Δxc, respectively) to ΔVASO/VASO. A MAGIC VASO sequence that provides whole‐brain coverage is demonstrated using a paradigm comprised of visual, motor, and auditory stimulation. Activated regions are quantitatively compared in the corresponding blood oxygenation level‐dependent (BOLD) images. Estimates of the minimum ΔCBV/CBV resulting from motor and visual stimulation were comparable to previous findings at 17 ± 8% (N = 8) and 19 ± 9% (N = 6), respectively. The absence of VASO activation for auditory stimulation and evidence of activation‐induced decreases in CSF volume fraction around the insula and superior temporal gyrus support the possibility of a Δxc contribution to the VASO signal. Without specific knowledge of the CSF components (xc,rest and Δxc), inference of ΔCBV/CBV from ΔVASO/VASO is severely limited. Magn Reson Med 58:306–315, 2007.

This publication has 24 references indexed in Scilit: