Cerebral blood volume mapping using Fourier‐transform–based velocity‐selective saturation pulse trains

• Published in: Magnetic resonance in medicine Vol. 81; no. 6; pp. 3544 - 3554
• Main Authors: Qin, Qin, Qu, Yaoming, Li, Wenbo, Liu, Dapeng, Shin, Taehoon, Zhao, Yansong, Lin, Doris D., van Zijl, Peter C.M., Wen, Zhibo
• Format: Journal Article
• Language: English
• Published: United States Wiley Subscription Services, Inc 01.06.2019
• Subjects: Adult; arterial spin labeling; Blood volume; Brain - blood supply; Brain - diagnostic imaging; Cerebral blood flow; cerebral blood volume; Cerebral Blood Volume - physiology; Coding; Computer Simulation; Defects; eddy current; Eddy currents; Female; Fourier Analysis; Fourier transforms; Fourier‐transform–based velocity‐selective saturation; Humans; Image Processing, Computer-Assisted - methods; Immunity; Magnetic Resonance Angiography - methods; Male; Mapping; Measurement methods; Middle Aged; Modules; Phantoms, Imaging; Robustness (mathematics); Saturation; Sensitivity analysis; Signal-To-Noise Ratio; Spin Labels; Substantia alba; Substantia grisea; Transit time; Velocity; velocity‐selective pulse train; eddy current; arterial spin labeling; cerebral blood volume; velocity-selective pulse train; Fourier-transform-based velocity-selective saturation
• ISSN: 0740-3194; 1522-2594; 1522-2594
• DOI: 10.1002/mrm.27668

Purpose Velocity‐selective saturation (VSS) pulse trains provide a viable alternative to the spatially selective methods for measuring cerebral blood volume (CBV) by reducing the sensitivity to arterial transit time. This study is to compare the Fourier‐transform–based velocity‐selective saturation (FT‐VSS) pulse trains with the conventional flow‐dephasing VSS techniques for CBV quantification. Methods The proposed FT‐VSS label and control modules were compared with VSS pulse trains utilizing double refocused hyperbolic tangent (DRHT) and 8‐segment B1‐insensitive rotation (BIR‐8). This was done using both numerical simulations and phantom studies to evaluate their sensitivities to gradient imperfections such as eddy currents. DRHT, BIR‐8, and FT‐VSS prepared CBV mapping was further compared for velocity‐encoding gradients along 3 orthogonal directions in healthy subjects at 3T. Results The phantom studies exhibited more consistent immunity to gradient imperfections for the utilized FT‐VSS pulse trains. Compared to DRHT and BIR‐8, FT‐VSS delivered more robust CBV results across the 3 VS encoding directions with significantly reduced artifacts along the superior‐inferior direction and improved temporal signal‐to‐noise ratio (SNR) values. Average CBV values obtained from FT‐VSS based sequences were 5.3 mL/100 g for gray matter and 2.3 mL/100 g for white matter, comparable to literature expectations. Conclusion Absolute CBV quantification utilizing advanced FT‐VSS pulse trains had several advantages over the existing approaches using flow‐dephasing VSS modules. A greater immunity to gradient imperfections and the concurrent tissue background suppression of FT‐VSS pulse trains enabled more robust CBV measurements and higher SNR than the conventional VSS pulse trains.