Dear Kevin,
Could you clarify what you mean by "transform the ROI to subject's space" with FLIRT?
Where does this ROI start off? And how is it defined?
If it is in standard space then I would say that this is a bad idea and you should use
the FNIRT warp to transform any coordinates or ROIs. If you define you ROI/coordinate
in some space that is individual to the subject, then a rigid alignment (with FLIRT) is
sufficient for registration.
If you can clarify we might be able to help more.
All the best,
Mark
On 1 Mar 2011, at 16:24, Kevin Denny wrote:
> We are doing seed-based resting state connectivity analyses. The way we convert the ROI coordinates to subject's native space is using FLIRT but the way we register each subject's anatomical to standard space is using FNIRT. We use nonlinear registration because we are using an elderly population, some with cognitive impairment, and the size of the ventricles and other structures tends to vary greatly in some subjects.
>
> Basically, I want to know if using FLIRT to transform the ROI to subject's space but using FNIRT in my analysis could cause any problems or if there is a better way to do this (i.e. use the same tool for both steps).
>
> Thanks in advance.
>
|