Figure 3. Design for the inter-modality statistical analyses.
First, we assessed the relationships between baseline measures of grey matter atrophy and the evolution of local hypometabolism, and between baseline measures of grey matter hypometabolism and the evolution of local atrophy using a voxel-to-voxel correlation approach (I and II and blue box). Second, we assessed the distant effects of each baseline alteration in predefined regions of interest (ROI) onto the evolution of alterations in other modalities using a ROI-to-voxel approach (III, IV, V and VI and yellow box).