Medical Physics Seminar – Monday, January 30, 2023
Advancing Functional Assessment for Cerebrovascular 4D Flow MRI
Grant Roberts
PhD Candidate | UW-Madison
4D flow MRI (time-resolved, volumetric, three-directional velocity-encoded MRI) is a powerful imaging method that can capture blood flow, vascular anatomy, and complex flow patterns in a single acquisition. In the brain, 4D flow MRI can be used for quantifying cerebral blood flow, velocities, and pulsatility in normal and neurovascular disease populations, blood flow tracking in arteriovenous malformations, risk assessment of aneurysm rupture using wall shear stress calculations, and more. However, proper quantitative assessment usually requires tedious image post-processing, including segmentation, centerline skeletonization, placement of analysis planes for flow, and flow visualization. The first part of this educational talk aims to present software tools that we have developed to automate cranial 4D flow post-processing, allowing for fast, robust, and repeatable flow visualization and quantification. In the second part of the talk, we apply these tools to measure cerebral blood flow and pulsatility in a large population (n=759) of older adults to establish normal flow and pulsatility baseline values across as large age spectrum.
Technical improvement of arterial spin labeling MRI for cerebrovascular evaluation in Alzheimer's disease
Mu-Lan Jen
PhD Candidate | UW-Madison
Alzheimer's disease (AD) accounts for over 60% of dementia cases and is the sixth leading cause of death in the US. While its mechanism remains unclear, recent studies have demonstrated that AD pathology is associated with neurovascular alternations, including decline in cerebral blood flow (CBF) and breakdown in the blood-brain barrier (BBB). Arterial spin labeling (ASL) is a non-contrast MRI technique that provides quantitative perfusion and permeability measurements to identify such microvascular alterations. Recent studies have shown promising results of using ASL to aid AD diagnosis. However, there are challenges in the signal-to-noise ratio (SNR), quantification error, and sensitivity to physiological variations and motion. This study aimed to (1) investigate and correct the sources of inaccuracy in ASL and (2) improve acquisition and reconstruction methods for ASL perfusion and permeability mapping.