Skip to main content
. Author manuscript; available in PMC: 2013 Mar 28.
Published in final edited form as: Neuroreport. 2012 Mar 28;23(5):283–289. doi: 10.1097/WNR.0b013e3283505b62

Figure 1.

Figure 1

Schematic diagram of the image normalization and quantification process. The images were normalized to a single-subject atlas, extensively parcelled in 3 dimensions[13]. For the analysis of diffusivity and white matter regional volumes, the normalization process was based on DTI contrasts (fractional anisotropy (FA) and b0 images, red labels), while, for the gray matter volumetric analysis, the normalization process was based on T1-WIs (blue labels). For the “forward” transformation, the subject images were normalized to the template first linearly, then by a highly accurate elastic algorithm (LDDMM). After this procedure, all brains have a shape similar to that of the atlas, and it is possible to compare subjects and controls voxel-by-voxel (voxel-based analysis - VBA). For the “backward” transformation, the parcellation map defined in the atlas is reverse-transformed to the original MRI. This enabled the 3-D automated segmentation of the original images and the structure-by-structure analysis (atlas-based analysis - ABA).