Quantitative T2: Software, application, workflows, phase correction.

摘要

Quantitative T2 (qT 2) is an emerging MRI field that utilizes the fundamental MRI T2 relaxation process and shows promise as a non-invasive biomarker for myelin health. This thesis addresses a few limitations of qT2 understanding and analysis, including both a thorough examination of qT2 attributes that characterize more than just myelin health and improved methods of analysis.;A histology study follows. Typically, qT2 is used as a myelin biomarker. However, the main outcome of qT 2 analysis is the T2 distribution, and the myelin water fraction, which is used as the myelin biomarker, is only a small part of the entire distribution. A multiple sclerosis human brain in vitro study was performed examining T2 distribution characteristics for regions defined by histopathological staining for myelin and axons. While it has been well established that the myelin water fraction is a myelin biomarker, this study finds that the position of another peak in the T2 distribution is affected by myelin and axonal loss.;The next study addresses the optimal qT2 analysis method. While drawing a region of interest allows the signal intensities within the region to be averaged together, thereby increasing the signal to noise ratio, the final measurement lacks estimates of error. Furthermore, simulations show that both averaging before performing the analysis and regularizing the analysis leads to undesirable changes to the T2 distribution. This study shows that a voxel-based, non-regularized analysis is the most robust way to perform qT2 analysis.;Finally, noise is examined. Quantitative T2 analysis is best performed on data with Gaussian distributed noise, but MRI noise is Rician distributed and simulations show that Rician noise affects the T2 distribution. A temporal phase correction method is developed and applied to in vivo human and rat qT2 data to change the noise from Rician to Gaussian prior to performing qT2 analysis.;To begin, a thorough introduction of qT2 in terms of acquisition and analysis is presented, followed by a review of applications. Four studies of qT2 follow. The first study addresses qT2 analysis, presenting an open source, platform independent qT2 software package that works with data independent of MR manufacturer.