Hi Clark, the orientation flags were not supposed to change the pixel data, but it turns out that the out_orientation will change the pixel data. I don't know how/when that happened.
doug
On 05/22/2012 03:56 PM, Clark Fisher wrote:
Hi everyone,
I have looked through the mailing list, and read a number of threads on using (or rather not using) mri_convert to change the orientation of a volume, and I have read the mri_convert help, which says about the --in_orientation and --out_orientation flags:
"This is only to be used when the orientation information in the input file is *wrong*. If it is correct, this will make it wrong!"
This lead me to believe that using these flags will only change the header, and not the data matrix, of my nifti files. However, I had a T1 volume in LIA format (taken from a monkey in sphinx position where mri_convert --sphinx had been applied), and needed to put it into RAS format for further processing. mri_convert --out_orientation RAS seemed to work for this, which fslswapdim (which I have seen recommended on the mailing list) introduced a Left/Right switch. That is to say, when I opened the volume output by mri_convert --out_orientation RAS with tkmedit, it displayed properly (unchanged from how it was before mri_convert), and when I loaded it into Matlab with MRIread, the .vol array (and not just the headers) had been reoriented. When I tried to use fslswapdim, on the other hand, the output volume had a left-right switch when viewed in tkmedit (I can tell thanks to a fiducial marker).
I'm confused by all of this: is mri_convert behaving like I would like it to, or am I just fooling myself? I'm happy to provide the volumes for other people to inspect. By the way, I'm using FreeSurfer v4.5.
-Clark _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer