Dear Bruce,
Some subcortical structures seems to be suboptimally segmented in our subjects. By looking at the original MRI intesities and labels of the atlas subjects individually we may suspect that the segmentation of these structures were also suboptimal in the atlas subjects (so the gca atlas is suboptimal), or it is rather related to some demographic/disease related differences between the training data and our subjects, and/or it is a segmentation algorithm related issue.
Best Regards, Gabor
2016-07-14 15:30 GMT+02:00 Gabor Perlaki petzinger.gabor@gmail.com:
Dear all,
Is it possible somehow to visually inspect the segmentations of the individual subjects used for the RB_all_2008-03-26.gca atlas? I can extract the means, priors and labels, but I am rather interested in the original MRI intensities and labels for the individual subjects.
Best Regards, Gabor