Q: Running nii2mnc under ubuntu gutsy, we are having trouble. We’re using spm5, and have tried saving as nii in one .nii file, and also as .img + .hdr file. When we try nii2mnc (on ubunutu), we get an error … can’t find the header. It won’t translate into mnc. What are we doing wrong?

A second question is whether there is right-left confusion on the nii2mnc conversions, or if the spm5 import takes care of it? We have some analyze format files (created using MRIcro, Chris Rorden’s program) that we put into spm5 to do normalization and segmentation, and now we want to get them back into David MacDonald’s register and Display programs running under ubuntu. We can get them into spm5, but not safely back again.
http://www.bic.mni.mcgill.ca/pipermail/minc-users/2008-June/002084.html

A:Ah, orientation.. :) First things first, MINC does not “do” radio vs neurological “format”, this was as abomination forced upon us by certain other packages who shall remain nameless. In short this this should be a viewing convention only. In the case of the BIC software all the tools that I know of view in neuro orientation. (register for one)

That said other packages do not do this and even go to the extents of flipping data on disk and defining global “orientation flags”. So in short I can guarantee nothing regarding orientation and files. The only way to be sure of this is to use a fiducial marker in an image and track it through your entire processing chain. mnc2nii follows the Nifti standard so I believe that a conversion to Nifti and then into FSL will not result in orientation issues.