Enrico
De Vita1,2, David L. Thomas2,3, Matthias Gnther4,5,
Xavier Golay1,3
1Lysholm Department of Neuroradiology,
National Hospital for Neurology and Neurosurgery, London, United Kingdom; 2Medical
Physics and Bioengineering, University College London, London, United
Kingdom; 3Institute of Neurology, University College London,
London, United Kingdom; 4Mediri GmbH, Heidelberg, Germany; 5Fraunhofer
MEVIS-Institute for Medical Image Computing, Bremen, Germany
In
arterial spin labelling (ASL) acquisitions, multiple post-labelling delays
(TI) are often used to quantify cerebral blood flow (CBF) without the
potential errors associated with assumptions on bolus arrival time. Q2TIPS
saturation is often additionally used to precisely define the temporal width
of the tagged bolus and facilitate CBF estimation. However the Q2TIPS pulse
train modifies tissue and blood signal via magnetisation transfer effects.
Here, we examine this undesired effect in multi-TI ASL with 3D-GRASE, in
particular how it alters the effectiveness of background suppression of
static tissue and therefore impacts on the SNR of the ASL measurement.