Measuring Cerebrovascular Pulsatility Using Cardiac Cycle Fluctuations of fMRI BOLD Data

摘要

Arterial pulsatility is linked to cerebral small vessel damage and neurodegeneration, but measuring cerebrovascular pulsatility in humans has largely been impeded by the skull. This thesis describes a method that generates cerebrovascular pulsatility maps based on resorting blood-oxygenation level dependent (BOLD) volumes according to their cardiac cycle position. Sensitivity of this method was tested using 20 minutes of moderate-intensity exercise as an acute physiological stressor in 45 healthy adolescents. Further examinations evaluated the influence of repetition time (TR) and echo time (TE) via simulation and multi-echo data, respectively. There were global, tissue-specific, and region-specific decreases in cerebrovascular pulsatility 20 minutes following exercise cessation. Cardiac-related pulsatility detection was comparable over a range of TR and TE values, with highest detection during rapid TRs (≤300ms) or shorter TE (~14ms). These results suggest that cardiac-related fMRI may represent a potent and easily adoptable method of mapping cerebrovascular pulsatility influences with voxel-wise specificity.