| dc.description.abstract | Conventional MRI is sensitive to detect brain abnormalities non-invasively through excellent contrast generated by variation in relaxation times and water proton density. However, conventional MRI lacks the specificity and quantitation to specific pathologies and tissue components such as myelin. Myelin is the major constituent of white matter ─an insulating macromolecular sleeve wrapped around axons of brain cells. Loss of white matter, specifically myelin leads to severe motor and cognitive deficits in diseases such as multiple sclerosis. Advanced quantitative MRI methods such as quantitative magnetization transfer (qMT) and diffusion tensor imaging (DTI) have emerged as putative biomarkers that improve sensitivity, specificity and provide quantitative metrics to measure myelin. qMT and DTI are model based quantitative techniques, which provide sub-voxel information of the underlying tissue architecture. qMT is sensitive to the tissue macromolecular content, whereas DTI is sensitive to tissue microstructure. Pool size ratio (PSR, a qMT parameter) and radial diffusivity (RD, a DTI parameter) provide an indirect quantitative measure of myelin. However, their relative sensitivities and specificities to myelin are unclear. qMT and DTI are based on different physical principles and may provide complementary information. While histology is the gold standard for myelin quantification, it can only be performed postmortem or through invasive biopsies. Thus, systematic quantitative MRI and histological validation studies are essential to determine the specific sensitivities of non-invasive quantitative metrics. Although, limited data is available on such studies due to their tedious, time intensive and complex nature. My thesis work addresses this gap by performing quantitative MRI and histological validation on a relatively new animal model of multiple sclerosis (MS), which recapitulates the inflammatory and non-inflammatory demyelinating phases seen in patients. The animal model was characterized using structural MRI, qMRI and histological methods. To enable quantitative comparisons amongst MRI and histological parameters, detailed processing protocol were designed and implemented including 3D qMT and DTI protocols and histological pipeline. In vivo and ex vivo studies were performed and qMT and DTI metrics were correlated with histology and among each other to determine their specific sensitivities. Furthermore, in an attempt to translate the animal work to clinical settings, a fast qMT sequence with GRASE readout was tested on human scanners. In conclusion, we found that PSR, and RD are sensitive to histological myelin content with PSR having the strongest correlation. | |
