Skip to main content
. Author manuscript; available in PMC: 2018 Feb 1.
Published in final edited form as: Neuroimage. 2016 Nov 12;146:100–112. doi: 10.1016/j.neuroimage.2016.11.020

Figure 7.

Figure 7

Results from the aging analysis in Sec. 3.2 showing a quantitative comparison of the methods. The plots show the R2 of linear regression models relating age to FA. The left plot shows results aggregated for each method and template type. The right plot shows results aggregated for each region and template type. Nine areas are shown: right anterior pericallosal white matter (R_PERI), the fornix (FORN), the superior and middle cerebellar penduncles (SCP/MCP), left uncinate fasciculus (LJJNC), splenium (SPLN), right superior frontal white matter (R_SUPF), right posterior thalamic radiation (R_PTR), and right inferior frontal white matter (RJNFF). Statistically significant diiferences (p ≤ 0.05) are marked with an asterisk. The results show that single voxel analysis performed better than region-based analysis. Skeletonization and smoothing did not significantly change performance, but the standard template performed better than the study template when used in conjuction with single voxel-based TBSS. There was moderate variability in performance across regions, and models of L_SCP and L_UNC were found to perform better with the standard template.