The problem is that mri_segstats does not recognize that 3414*3=10242. You'll have to convert the nifti file to change dimensions. You can do this with mri_surf2surf. Specify fsaverage5 as the source and target subject. The output should have the correct dimensions. doug

Linda Douw wrote:

Hi all,

I have nifti files containing the surface-projected time series data of a subject (obtained using procsurffast), both lh and rh. These time series are projected onto fsaverage5, and are 3414 x 1 x 3 time points, where 3414*3 = 10242 are the number of vertices for each hemisphere. Now, I would like to extract the mean time series for 138 ROIs, which are defined in an annotation called sparc I made:

mris_divide_parcellation fsaverage5 lh aparc.annot sparc.txt lh.sparc

I would assume that these files are now of the same dimensions. However, when I run:

mri_segstats --i /subject/surf/subject_projectedto_fsaverage5.nii.gz --avgwf /subject1/ts_lh.txt --annot fsaverage5 lh sparc

it crashes because the dimensions are not the same (the segmentation is 10242 x 1 x 1). They do both overlay nicely on fsaverage5 inflated using tksurfer.

What may cause this mismatch? And is there a solution?

Thanks, Linda

---

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