Method for Optimizing Quantitative Temporal Lobe Epilepsy MR Imaging

Date
2018-01-17
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Many neurological disorders such as epilepsy rely on MRI for detection of structural abnormalities. However, the current clinical MRI methods are insufficient and insensitive in detection of subtle abnormalities. MRI quantitative T2 mapping is a promising quantitative medical imaging technique as it is highly sensitive to tissue composition. The conventional approach for T2 mapping assumes mono-exponential signal decay; however, this is rarely observed due to transmit field inhomogeneity and miscalibration at high field MRIs. The nonexponentially results in poor fits and a systematic bias in estimated decay rates. A recently proposed fitting method, called stimulated echo correction, uses the same input data but estimates the major confounds associated with mono-exponential fitting. Optimal accuracy and non-optimal precision is achieved in this method. My first aim was to develop a stimulated echo correction based method with fewer parameters and higher precision relative to the original one. The second aim was to implement this new method in order to better identify abnormal brain regions in temporal lobe epilepsy that were poorly visualized on standard images. I hypothesized that my improved stimulated echo correction with fewer parameters would provide more accurate and reliable transverse relaxometry imaging than does conventional or the original stimulated echo correction fitting methods, and would improve our ability to detect subtle irregularities associated with epilepsy. The new method was evaluated with simulated and in-vivo data, in which up to 27% reduction in variance in the new method compared to the original stimulated echo correction was observed. Moreover, the new method had greater reliability in categorizing abnormalities in hippocampal regions when compared with exponential and stimulated echo correction methods. I concluded that the new method is able to reduce the variance in T2 relaxometry from multi-echo spin echo sequences; therefore, this method can potentially help in detection of those lesser obvious hippocampus abnormalities.
Description
Keywords
MRI, Medical Imaging, Relaxometry, T2, Epilepsy
Citation
Basiri, R. (2018). Method for Optimizing Quantitative Temporal Lobe Epilepsy MR Imaging (Master's thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca. doi:10.11575/PRISM/5418