Hi David,
we haven't sorted out all the non-1mm issues yet. The easiest (and admittedly annoying) thing to do is to use mri_convert to change the effective voxel resolution to 1mm. That is, tell it that they are 1mm without changing the data. Then you'll need to remember to rescale all your final results like thickness and volume.
Sorry, this is high on our list of things to fix.
Bruce
On Tue, 20 Mar 2007, David M Schnyer wrote:
Folks,
We have a .7mm cubic voxel SPGR scan we would like to segment with Freesurfer. When I run it with the recon all flag -cm images retain their "native" resolution up to the normalization step of nu to T1. nu.mgz has the native resolution but T1 is converted to 256 cubic. Can I do what I want here?
thanks,
david