Electronic Thesis and Dissertation Repository


Master of Science


Medical Biophysics


Dr. Mel Boulton


Intracranial vessel wall imaging may be accomplished with high-field (7T) magnetic resonance (MRI). To determine its feasibility, a 7T MR protocol was defined using Polyvinyl-alcohol cryogel (PVA-C) phantom vessels and healthy subjects.

A 2D matrix construct of PVA-C vessel phantoms of different diameters and wall thicknesses was scanned. Three observers measured the phantom images, one of which three times. Physical measurements were performed using a digital caliper. Ten volunteers were scanned using three different MRI sequences (TSE-3D, FLAIR, MPRAGE). Imaging assessment was performed in different circle of Willis (COW) segments. Reliability and accuracy of the measurements was analyzed by inter and intraobserver correlation and by comparison to physical measurements.

Phantom measurements showed overall high inter and intraobserver reliability and accuracy (ICC≅0.9). However, precision diminished for smaller vessels (<3mm). TSE was superior on vessel wall definition compared with FLAIR on both, phantoms and volunteers. On healthy subjects, vessel wall was recognized consistently, but precise definition of distal COW segments was not achieved. Vessel wall was significantly overestimated (p<0.05) when comparing to intracranial vessel diameters from prior studies due to partial volume effects.

Vessel wall imaging is feasible with 7T MR. However, precision and definition decreases consistently with the vessel caliber. PVA adequately mimics 7T MR vessel wall imaging properties.