Hi Aimi, this is not possible with nifti since 163842 > 32768 (2^15) which is the limit for nifti. It should have worked with mgh output and --noreshape. Can you confirm that it does not? doug

On 10/10/2012 11:15 AM, Aimi Watanabe wrote:

Hello,

My input to freesurfer is:

mri_surf2surf --cortex --fwhm-src 5 --noreshape --hemi rh --tval [$output file] --sval [$input file] --srcsubject [$subject id] --trgsubject fsaverage --tfmt nii

When I reopen the resulting image with Nibabel, the dimensions are 27307 x 1 x 6 x 296. While 296 is the time axis, I was hoping to get 27307*6 = 163842 vertices.

I have replaced '--noreshape' with '--reshape', '--reshape_factor=6'

and the bash output looks proper:

121010-16:49:28,430 interface INFO: stdout 2012-10-10T16:49:28.429912:Saving target data 121010-16:49:28,430 interface INFO: stdout 2012-10-10T16:49:28.429912:Reshaping 6 (nvertices = 163842) 121010-16:49:28,430 interface INFO: stdout 2012-10-10T16:49:28.429912:Saving to [$output file]

However, it does not change the resulting image.

I have also tried other file types that are not 'nii', like 'mgh', but this does not help either.

Please advise?

Also, I could easily make a work-around, and reshape the image from a(27307, 1, 6, 296) to b(163842, 1, 296) in FORTRAN (column-major) order. I.E.: 0~27306 then add the next set 27307~54613. If I do this, would the original vertex map be preserved?

Thank you! _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer