Velocity Selective Arterial Spin Labeling using an Adiabatic Hyperecho Pulse Train

E. C. Wong, M. Cronin UCSD, La Jolla, CA, United States Synopsis In Velocity Selective Arterial Spin Labeling (VS-ASL), arterial blood is magnetically tagged using velocity selective pulses with no spatial selectivity, resulting in small and uniform transit delays throughout the region of interest. We introduce here the use of a hyperecho based tagging pulse train that consists of BIR-4 adiabatic rotation pulses with gradient pulses interspersed, followed by a delay to allow for inflow and then image acquisition. This pulse train saturates flowing blood with improved velocity selectivity, B1 insensitivity and off-resonance insensitivity, and also leaves static tissue magnetization inverted for background suppression. Introduction In the ideal case, the tagging pulse for VS-ASL generates a rectangular profile of Mz vs flow velocity. A sharp cutoff velocity allows for accurate quantitation of tissue perfusion, and a flat response below the cutoff velocity provides insensitivity to slow physiological motion (1). In (1) the velocity selective tagging pulse consisted of a 90x-grad-180y-grad-90-x pulse train that generates a profile that is sinc shaped in the presence of laminar flow. Methods In order to improve the velocity selectivity profile we explored (αi-gradi-180-gradi)n-αn+1 trains as suggested by Norris (2), and also hyperecho (3) based trains using Bloch equation simulations, and found that the hyperecho based trains are more time efficient. This is presumably because the off-resonance insensitivity is provided by the single 180° pulse at the center of the train, as opposed to multiple 180° pulses. Velocity selective hyperecho pulse trains of the form (αi-gradi)n-180-(-αn+1-i-gradn+1-i)n were designed using the Nelder-Mead simplex method to optimize the αi, their phases, and the area of the gradients. The error function was the mean squared difference between the velocity profile and an ideal rectangular profile with a velocity cutoff of 1cm/s. For B1 insensitivity, BIR-4 pulses (4) with tan(κ)=3, ζ=3 and width=6ms were used for all pulses. For optimizations, T1 and T2 were assumed to be 1000ms and 100ms, respectively, and pulse trains with n=2 (5 RF pulses total) were found to be optimal, as improvements in the shape of the profile with higher n were offset by increased T2 decay during the pulse. Unlike the applications outlined in (3), the hyperecho pulse train did not start with a 90° pulse, as this would unnecessarily reduce the degrees of freedom for the optimization. If the final RF pulse is set to –α1 (with phase –φ1), then static magnetization will be left inverted. If it is set to 180°-α1 (with phase –φ1), then static magnetization will be returned to the +Z axis. Either of these can be used for VS-ASL, as each results in the same difference in Mz between static and flowing spins. The inverted case naturally results in at least partial suppression of the static tissue signal, resulting in reduced noise in the time series, and also reduces the dynamic range of the signal, which can be useful for high dynamic range imaging methods such as volume acquisitions. Results An optimized hyperecho pulse train is shown in Figure 1. Parameters of this pulse are: α = {58.2,-94.1,180,94.1,-58.2}°, φ = {86.0°,-113.6,0,113.6,-86.0}°, gradient durations are {0.79,2.37,0.79,2.34}ms, using trapezoidal gradients of 3G/cm amplitude and ramps slewed at 14G/cm/ms. The calculated velocity profile for this pulse is shown in blue in Figure 2 for the non-inverted pulse train. The sinc profile of a spin echo VS pulse is shown for comparison. Note the improved sharpness of the velocity cutoff, as well as the flatter profile near zero velocity. An example VS-ASL image acquired in 200s on a Varian 3T scanner using a single shot EPI readout is shown in Figure 3.