Tensor-based morphometry is a somewhat different method that is especially suitable for mapping growth rates for different structures. In TBM (Gogtay, 2008; P. M. Thompson et al., 2000), 3D maps of local tissue growth rates (as a percent per year) are derived by having a computer algorithm fluidly reshape the earlier scan to match the later one, matching fine-scale anatomical features. The applied deformation is analyzed to create per-subject maps of the growth rate (expansion factor) or local loss rate of tissue. In this approach, there is no need to quantify the amounts of gray and white matter in the scans, nor is it required to hand-label anatomical structures on the scans, unless summaries of growth rates are needed for particular regions of interest. The maps of growth rates for each subject can then be aligned to match a common anatomical template. As in VBM, statistics can be compiled and group differences in growth rates, or effects of medication, risk genes, or covariates of interest, can be plotted at each location the brain. In recent longitudinal MRI studies, TBM has
