MNI-FTD templates, unbiased average templates of frontotemporal dementia variants

Prior to generating the average templates, the quality of the images (e.g. presence of image artifacts such as motion) as well as the linear and nonlinear registrations was visually assessed and 16 cases (2 CN, 7 bv FTD, 3 svFTD, and 4 pnfaFTD) that did not pass this quality control step were discarded. Similarly, 10 cases (4 CN, 4 bv FTD, 1 svFTD, and 1 pnfaFTD) failed quality control for tissue segmentation step and were discarded before generating the tissue probability maps. Figures S6–S8 in the supplementary materials show examples of QC images for cases failed due to presence of motion artifact (Figure S6), nonlinear registration failure (Figure S7), and tissue segmentation failure (Figure S8). For comparison, Figures S9, S10, and S11 show examples of passed cases for linear and nonlinear registrations and tissue classification, respectively. No cases failed quality control…