You are correct, surf2surf is not changing the values from out side the masked region. However the problem is that in smoothing the values inside the masked region surf2surf appears to be incorporating values from outside the masked region (even if its not smoothing these "outside" values).

What I would like to do is to restrict surf2surf to smoothing only within the mask and not use any values that lie outside of it.

I apologize if I'm being unclear, I'm at a bit of a loss for out to accurately describe this.

Thanks, Greg

On Thu, 14 May 2009, Douglas N Greve wrote:

mri_surf2surf will not change the values outside of the mask (those will be left unsmoothed). Is there an actual difference in the mask?

doug

Gregory Dierksen wrote:

...

Hi,

I want to smooth a cortical thickness map using values only in a specific area. Basically I want to mask out all other segmentations, for example smooth with fwhm of 20 only in the postcentral region.

I've used quite a few combinations of the -label-src and -label-trg flags and noticed a slight problem:

Using the command line (which seems to give the "best" results):

mri_surf2surf --hemi rh --s fsaverage --sval ./rh.Input.mgh --fwhm 20 --label-trg ./postcentral.label --tval rh.Output.mgz

When the rh.Input.mgh has been masked before running surf2surf using postcentral.label (so only the postcentral region is non-zero), I get DIFFERENT values than if rh.Input.mgh is NOT previously masked by postcentral.label (ie ALL verticies have non-zero values inside and outside postcentral region).

How can I get mri_surf2surf (or perhaps some other command) to smooth only within a specified region without taking values from outside this region?

Thanks for the help,

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

--

I'm working with a Stable 4 version of FreeSurfer.

I think your explaination about the vertices is acctually the answer to my question. The variation that I'm seeing between the pre-masked, and non-pre-masked surface is quite small, but never-the-less present. If this is the case, I think the current mathed I have will suffice for out purposes.

Thanks a lot for your help.

Greg

On Thu, 14 May 2009, Douglas N Greve wrote:

What version are you using? I just checked this on my current version (and this part has not changed in a long time), and it gave the same result in the mask regardless of whether I masked the input or not. One thing that I did find is that it will smooth the vertices that are outside the mask but immediately adjacent to it. This should be a small effect but it might be misinterpreted as differences inside the mask.

doug

Gregory Dierksen wrote:

...

You are correct, surf2surf is not changing the values from out side the masked region. However the problem is that in smoothing the values inside the masked region surf2surf appears to be incorporating values from outside the masked region (even if its not smoothing these "outside" values).

What I would like to do is to restrict surf2surf to smoothing only within the mask and not use any values that lie outside of it.

I apologize if I'm being unclear, I'm at a bit of a loss for out to accurately describe this.

Thanks, Greg

On Thu, 14 May 2009, Douglas N Greve wrote:

...

mri_surf2surf will not change the values outside of the mask (those will be left unsmoothed). Is there an actual difference in the mask?

doug

Gregory Dierksen wrote:

...

Hi,

I want to smooth a cortical thickness map using values only in a specific area. Basically I want to mask out all other segmentations, for example smooth with fwhm of 20 only in the postcentral region.

I've used quite a few combinations of the -label-src and -label-trg flags and noticed a slight problem:

Using the command line (which seems to give the "best" results):

mri_surf2surf --hemi rh --s fsaverage --sval ./rh.Input.mgh --fwhm 20 --label-trg ./postcentral.label --tval rh.Output.mgz

When the rh.Input.mgh has been masked before running surf2surf using postcentral.label (so only the postcentral region is non-zero), I get DIFFERENT values than if rh.Input.mgh is NOT previously masked by postcentral.label (ie ALL verticies have non-zero values inside and outside postcentral region).

How can I get mri_surf2surf (or perhaps some other command) to smooth only within a specified region without taking values from outside this region?

Thanks for the help,

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

--