Hi,

I want to run some surface analysis on resting state data. So, First I used Freesurfer to extract the surfaces and applied SUMA on them, having standard meshes ready. Also I did common functional data preprocessing steps, and finally used 3dVol2surf and Surfsmooth to smooth the data as follows:

3dVol2Surf \
-spec ${Subject}_surface/SUMA/std.141.${Subject}_surface_${hem}.spec \
-surf_A std.141.${hem}.smoothwm.asc \
-surf_B std.141.${hem}.pial.asc \
-sv ${Subject}/${Subject}_surface_SurfVol_Alnd_Exp+orig \
-grid_parent ${Subject}/rest_res+orig. \
-map_func ave \
-f_steps 10 \
-f_index nodes \
-out_niml ${Subject}/rest_res.${hem}.niml.dset -overwrite

SurfSmooth -npb 1 -overwrite \
-spec ${Subject}_surface/SUMA/std.141.${Subject}_surface_${hem}.spec \
-surf_A std.141.${hem}.smoothwm.asc \
-met HEAT_07 \
-blurmaster ${Subject}/rest_res.${hem}.niml.dset -bmall \
-target_fwhm 6 \
-input ${Subject}/rest_res.${hem}.niml.dset \
-output ${Subject}/rest_res6.${hem}.smoothed.niml.dset

But the niml.dset smoothed functional output of Surfsmooth number of nodes differs by standard meshes number of nodes. I used std.141, so the number of nodes are 198812, but for example in a subject the number of nodes in smoothed.functional.niml.dset is 198726. And it varies subject to subject in most of cases, only 40% of the subjects' functional data have the same number of nodes as standard meshes. I should say that the functional data seems to be missed in some boundaries (which I guess it's the reason), but could you please let me know if it is the reason and how to fix it or at least know which nodes are missed automatically?


Thanks very much for your help.
Mahshid