Hi Shiran
if you have the FA registered why do you need to recompute the transform for the other volumes? They are all derived from the same underlying data so should all be in perfect register with each other and hence should share a registration transform.
cheers Bruce
On Thu, 16 Mar 2017, Su, Shiran wrote:
Hi, FreeSurfer developers, I am trying to use the command 'bbregister' to register the axial diffusivity, mean diffusivity, radial diffusivity and FA maps to the segmentation result (get from the 'recon-all' command). The registration works well for the FA map. But for AD, MD and RD maps, the registration is a little off (as can be seen in the example). Although we could use the 'transform volume' tool (tkregister tool) to move the diffusion maps to make the map fit better with the segmentation surface, I am wondering whether this is a normal issue and we could only use 'transform volume' tool to modify them? Because our AD, MD, RD and FA maps are all masked and have good location correspondence with each other. If not, is there any method we could use to get a better registration result for the AD, MD and RD maps?
i.e. [IMAGE][IMAGE]The registration works well for the FA map (left) but not for the AD map (right).
Thank you!
Best, Shiran Su