Item

Abstract

Purpose/Introduction: Continuous arterial spin labelling is the most sensitive
technique for perfusion measurements. Several schemes for tag and
control scans have been developped to overcome the technical and condeptual
problems of this technique. Here, four different approaches for tag and
control are simulated and compared experimentally.
Subjects and Methods: Four sequence variants are compared that differ
only in the way the tag or control pulses are applied: In CASL, separate tagging
coils at the neck label the inflowing blood with one long pulse [1]. In
ACASL, the tag is applied by the same coil used for imaging [2], but, due to
technical constraints, this pulse has to be interrupted at regular intervals.
For multi-slice acquisition (ms-ACASL) the control scan uses an oscillating
pulse to avoid nonsymmetric magnetization transfer effects [3]. In PCASL
[4], the tag is composed of a large number of short pulses and gradient
shapes, with the control scan modifying the puls phases as to be transparent
to the flowing spins. These four tagging techniques have been simulated and
implemented on a Siemens Trio with an EPI-readout. The simulations were
used to optimize the parameters of all sequences.
Results: Two volunteers were examined with each of the four techniques, as
well as with a standard FAIR sequence as representative of pulsed ASL. For
all CASL measurements, the tagging duration was 2 s; with a delay of 1s before
acquiring 7 slices (ss-ACASL: 1 slice). Only the center slice was used for
futher analysis. Figure 1 shows the perfusion images as generated with FSL
from the same slice in one volunteer acquired with all five sequences. Voxels
with a significant perfusion signal (Z = 3.7) are red. Sensitivity differences
between the sequences result in a different number of significant voxels.