MRS-based classification of Spinocerebellar Ataxias 1, 2, 3, and 6 at 7T: A distance-weighted discrimination approach
James Joers1, Dinesh Deelchand1, Tianmeng Lyu2, Uzay Emir1,3, Diane Hutter1, Christopher M Gomez4, Khalaf O Bushara5, Lynn E. Eberly2, and Gulin Oz1
1Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, MN, United States, 2Division of Biostatistics, University of Minnesota, Minneapolis, MN, United States, 3University of Oxford, Oxford, United Kingdom, 4Department of Neurology, University of Chicago, Chicago, IL, 5Department of Neurology, University of Minnesota, Minneapolis, MN
We investigated the sensitivity of ultra-high field MRS to distinguish 4 hereditary neurodegenerative diseases with substantial overlap in clinical presentation and conventional MRI. We carried out pairwise classifications of spinocerebellar ataxias (SCAs) 1,2,3 and 6 using distance weighted discrimination (DWD) on 7T MRS from 3 brain regions of 68 SCA patients. Each subject contributed to the DWD model a 50-dimensional vector of concentrations. SCA6 was classified from SCAs 1-3 with high reliability (90%) and SCA2 was classified from SCA3 with moderate success (84%), while SCA1 could not reliably be classified from SCA2 (62%) or SCA3 (48%) with the current model.