who knows. could be cool though to train a neural network to recognize specific folds or something in an atlas, and then automatically segment an image. seems to me as atlas size increases your chances of actually including voxels from a specific target region of the brain will decrease due to minute differences in subject-specific registration, so this could also facilitate a cool investigation on that side of things.
who knows. could be cool though to train a neural network to recognize specific folds or something in an atlas, and then automatically segment an image. seems to me as atlas size increases your chances of actually including voxels from a specific target region of the brain will decrease due to minute differences in subject-specific registration, so this could also facilitate a cool investigation on that side of things.