These regions tend to be pretty well aligned, so you probably need a
bit less than you would for the cortex. However, you may need to use
some smoothing to suppress some of the aliasing effects that occur.
These depend on the resolution of your original scans. At a rough
guess, I'd suggest that somewhere between about 4mm and 8mm could be
reasonable, although I don't have any evidence to back this up.
Best regards,
-John
On 10 February 2011 19:38, Meng-Chuan Lai <[log in to unmask]> wrote:
> Dear John,
>
> Thank you for the very valuable information.
>
> With these in mind, practically, if to run VBM and we expect to find regional differences in small areas in the brain such as hypothalamus, ventral striatum or amygdala, what would be the lowest effective smoothness for DARTEL-processed modulated images with 1mm isotropic voxels, that the registration error should have been taken care of by such smoothness?
>
> Thank you!
>
> best regards,
> Meng-chuan Lai
>
|