Meeting Banner
Abstract #0820

Combining SIENA & SIENAx for Improved Quantification of Grey and White Matter Atrophy

Mishkin Derakhshan1, Sridar Narayanan1, D. Louis Collins1, Douglas L. Arnold1

1Montreal Neurological Institute, McGill University, Montreal, Quebec, Canada

We propose a novel technique for a more sensitive measurement of tissue-specific atrophy by combining the precise, longitudinal edge-displacement approach of SIENA with the tissue classification feature of SIENAx. The method is evaluated using (1) scan-rescan data and (2) simulated atrophy data and (3) is applied to a multi-centre relapsing-remitting multiple sclerosis dataset.

Keywords

absolute accounted accuracy accurate accurately achieved addition advantage affected aging amount appear appears applicable assess assessment assist atrophied atrophy automated avoid better brain causing certain classification clinical clinically combining common compartment complete comprises computing correcting cortical counterpart creating cross currently dataset dependence detecting detection dimensions diseases distortion distortions driven entirety error estimating example extraction feature fluctuations fraction global gradient hand illustrated impact importance improvement independently largely lesion limitations linear longitudinal masking measure measures much multiplying normalization normalized option output pair particularly people percent positioning practical precise problem proposed qualitatively quantification rather reduce reduced relapsing relevant relies remitting repeated reported reproducibility reproducible rescan resolution respectively robust scaled scaling scanned scanner sclerosis sectional segmentation sensitive sensitivity separate serial significance significantly simply simulated since slice slices slightly smith still studies subject subjects subsequent template terms tissue tissues towards transform treating trial trials typically understanding unknown version vital volume volumes white whole widely worse years