1 mmoL/kg).

DSC scan parameters included echo times (TE) ranging from 23 to 50 ms, repetition times (TR) ranging from 1,250 to 1,400 ms, flip angles (FA) ranging from 30 to 35 degrees, 40 to 90 repetitions (temporal time points), slice thickness ranging from 4 to 7 mm with interslice gap ranging from 0 to 1.5 mm, number of slices ranging from 6 to 20, and matrix size ranging from 80 × 96 to 128 × 128, depending on whether perfusion data were acquired on a 1.5 T or 3.0 T system. Data analysis of DSC data was performed offline using commercially available postprocessing software (IB Neuro v2.0™; Imaging Biometrics, LLC, Elm Grove, WI, USA). DSC analysis consisted of the following steps: (1) truncation of the first five time points in the DSC time series, since the MR signal does not reach steady state before this time, find more (2) calculation of the prebolus signal intensity on a voxel-wise basis, (3) conversion of truncated DSC time series to a concentration-time curve based on the T2* relaxivity of the contrast agent, and (4) estimation of CBV on a voxel-wise basis by using a 120 point trapezoidal integration

with correction for leakage, as described in previous publications.[2, 8-10] CBF was calculated using circular deconvolution of the arterial input function, which was chosen automatically in five voxels using IB Neuro v2.0™. An inherent constraint to 2D ASL acquired using echoplanar acquisition is the limited number of obtainable images, reducing the amount of total brain coverage. Additionally, each slice acquired with 2D ASL experiences slightly different Vincristine nmr inflow time, thus it is difficult to estimate a precise transit time when multiple slices are acquired. The use of 3-dimensional acquisition techniques overcomes many of these limitations, allowing both whole brain coverage and simultaneous acquisition to ensure a unified mean transit time. Pseudocontinuous ASL provides the main advantages of pulsed ASL, including a slightly lower radiofrequency

power deposition and higher inversion efficiency, while maintaining the benefits known to continuous ASL, such as the ability to tag spins within a physiological MCE range of velocities and higher overall tagging efficiency. In the current study, ASL PWI scans were performed using a PCASL pulse sequence with background suppressed 3D GRASE (Gradient and Spin Echo) readout (postlabeling delay = 2 second, FOV = 22 cm, matrix = 64 × 64, 26 × 5 mm slices, rate-2 GRAPPA, TR = 4 second, TE = 22 ms, 30 pair of tag and control acquired in 4 minute)[11, 12]. Data analysis was performed with Interactive Data Language (IDL (Boulder, CO, USA)) software programs developed in-house. ASL images were corrected for motion, pairwise subtracted between label and control images followed by averaging to generate the mean difference image (ΔM).