Quantitative Cerebrovascular Magnetic Resonance Imaging
Date
2014-09-19
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
This thesis explores quantitative cerebrovascular magnetic resonance (MR) imaging, a broad topic, with the aim of providing relevant numerical values associated with blood flow through the brain. Anatomy, pathology and basic angiography methods were reviewed. Several other MR imaging methods for obtaining cerebrovascular measurements are reviewed.
Exploration of the lowest achievable variance with MR imaging was undertaken through simulation using a digital brain phantom. A phantom was constructed from a healthy human brain data set using advanced methodologies to yield volumes of MR parameters (i.e., coil sensitivity, B0, B1, M0, T1, T2, T2*, and magnetic susceptibility). The digital brain phantom was then used to simulate the MR acquisition process and generate images, in order to determine the minimal achievable variance as a function of coil profile distortion. It was found that the degree of coil correlation could affect the lowest achievable variance by up to 2× to 3× over practical ranges.
The focus of the experimental chapters is on phase contrast velocity mapping and metrics that can be derived from velocity maps, such as: peak velocity, volume flow rate, and intravascular pressure. Prospective imaging was performed on healthy humans, and eight patients (five cerebral aneurysms and three arteriovenous malformations). A case study of a giant cerebral aneurysm was explored in greater detail, and stent treatment was shown to reduce flow asymmetry. Peak velocity and volume flow rate was determined for vessels in the normal brain. Bootstrapping is performed to assert that group-wise measurements are representative of the broader population and flow laterality is examined. Significant flow asymmetry was found between several paired vessel segments. Flow in the patients was imaged, and derived metrics were compared to the healthy cohort. Patients with aneurysm were found to have significantly lower flow in vessels distal to the aneurysm, while arteriovenious malformation patients were found to have significantly higher flow in vessels supplying the nidus.
Description
Keywords
Engineering--Biomedical
Citation
MacDonald, M. E. (2014). Quantitative Cerebrovascular Magnetic Resonance Imaging (Doctoral thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca. doi:10.11575/PRISM/27505