t1_class.nii.gz and t1_class.nii files in vistadata have incorrect headers

[j-ales]

The class files in the vistadata/anatomy and vistadata/functional directories have headers that are incorrect and results in an incorrect orienting of class file with the T1's when the header information is actually used. Current class reading code doesn't read the header information so it didn't matter. But soon the class reading code should be harmonized with the new anatomy reading code.

[JWinawer]

Yes, this is true and a good catch. The only reason things work ok now for many of us is that we happen to have two error that cancel each other out. They both need to be fixed at the same time, as you say.

The problem is that the t1_class files are typically made with a freesurfer call that is not correct. Simultaneous with this fix, we need to fix the way vistasoft reads the class files, which I think you did on your git branch. Sometime next week I will fix the sample data sets on vistadata, change the freesurfer call we use to make the class files, and merge your fixes to readClassFile. Hopefully all will work out ok.

Jon

On Jul 26, 2013, at 7:02 PM, j-ales wrote:

The class files in the vistadata/anatomy and vistadata/functional directories have headers that are incorrect and results in an incorrect orienting of class file with the T1's when the header information is actually used. Current class reading code doesn't read the header information so it didn't matter. But soon the class reading code should be harmonized with the new anatomy reading code.

—
Reply to this email directly or view it on GitHub.