Voxel-Based Multiple Regression of Multimodal MRI: Applications to Physiological Aging

Andrea Cherubini¹, Patrice Pran¹, Carlo Caltagirone¹, Gianfranco Spalletta¹

We explored for the first time with a voxel-based approach the simultaneous variation induced by physiological aging on four quantitative MR parameters sensitive to complementary tissue characteristics (VBM, T2* relaxometry, DTI). This allowed us to compare the performance of different predictors and to identify without a priori information the best biomarker of age-induced structural variation for each voxel. Our results showed that brain areas most affected by age are evenly distributed between white matter and grey matter. Moreover, the best quantitative predictors in most brain areas resulted to be iron deposition and microstructural damage rather than macroscopic atrophy of tissues.