Files
Abstract
Magnetic resonance imaging (MRI) is a non-invasive medical imaging technique that extends well beyond its visual, qualitative applications. Through the manipulation of quantum mechanical spin, which is the basis for MRI, quantitative information about the intrinsic physical and chemical properties of tissues can be determined. In this dissertation, multiple different MRI techniques are employed in order to estimate various properties of tissues, assess different methodologies of estimation, and discuss practical applications of these techniques. Among the techniques utilized includes: triglyceride composition mapping and quantitative susceptibility mapping (QSM) using chemical shift encoded (CSE-) MRI, functional (f) MRI, and diffusion tensor imaging (DTI). Part 1 of this dissertation focuses on the use of CSE-MRI in fatty tissues to quantify triglyceride composition and liver iron concentration (LIC). Many methodologies for estimating triglyceride composition require an a priori fat spectral model; therefore, the effect of using various different fat spectral models on this estimation is explored in Chapter 2. It was observed that the triglyceride composition had a strong dependence on the chosen a priori model; however, proton density fat fraction was independent of the chosen model. The primary MR-based methodology for determining LIC is currently the effective transverse spin-spin relaxation time (R2*) mapping, whereas QSM is a newer methodology that may have benefits over R2* mapping. Chapter 3 compares the use of R2* mapping and QSM to quantify LIC in an ex vivo mouse model. QSM proved to be a more robust methodology in this ex vivo study. Part 2 of this dissertation focuses on using fMRI to study pig brain functional connectivity and the disruptions caused by traumatic brain injuries (TBIs). Due to the many similarities between the pig and human brain, the pig serves as a useful large animal model for studying brain connectivity. Chapter 5 establishes six resting-state networks (RSNs) in the pig brain, draws comparisons with the human brain, and uses quantitative DTI measurements to provide support of the validity of these RSNs. Chapter 6 seeks to detect and evaluate disruptions in these RSNs using a TBI pig model. Disruptions were successfully detected in four RSNs and were traced back to affected individual anatomical regions.