Dear Paul,

mris_expand can be used only in the hypothetical cases, where in all brain systematically the surface is shifted to equal amount from the position which is anatomically relevant.

Therefore it cannot be used to correct for local errors. To ilustrate the use of this tool, see the screenshot where I tried to use this tool to "move" white surface to the half of the thickness towards the pial surface. Red is original white surface, yellow is original pial surface, blue is the surface produced by mris_expand.

The error you showed in the screenshot (part of the gyrus not encircled by pial and white surface) is usually corrected by placing control points to the white matter, see the tutorial:

https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/ControlPoints_freeview

By correcting the white surface this way usually also pial surface is moved to correct position.

Regards,

Antonin

Hello Antonin, GM/WM boundary = thickness. correct me if i am wrong. Shouldn't expanding/shrinking the ?h.white or ?h.pial surface to the correct size and shape help with that? I have moved the position of the ?h.white to capture most of the WM using mris_expand but i haven't been successfully with ?h.pial using mris_expand because the input and output were identical. Hence, I am looking for a way to do it with ?lh.pial and how to tell recon-all to incorporate my changes. Please correct me since I am still new to this (i.e., using individual recon-all scripts). Thank you. Best, PaulOn Wed, Jan 11, 2017 at 12:32 PM, Antonin Skoch a...@ikem.cz wrote:

Dear Paul,

I am quite surprised that you were successful to estimate correct GM/WM boundary using this tool. As far as I know and looking at the source code, this tool GLOBALLY moves the position of the specified surface to some amount, so it cannot be used to correct local errors. It is used e.g. for estimating the surface ad the midthickness of the gray matter.

See e.g. https://mail.nmr.mgh.harvard.edu/pipermail/freesurfer/2009-July/011152.html

Antonin

Thanks bruce for your response. I am trying to expand the pial surface (see attached image) in order to estimate the GM/WM boundary properly. I have successfully expanded the white matter but when i tried the mris_expand scripts for pial surface (mris_expand -thickness lh.pial 0.5 lh.newpial), the output looked identical to the input. Please can you direct me on how to successfully achieve this for the pial surface and how to incorporate my changes (expanded surfaces) into recon-all. Thanks. Best, Paul

On Wed, Jan 11, 2017 at 11:37 AM, Bruce Fischl fis...@nmr.mgh.harvard.edu wrote:

...

Hi Paul

it's hard to answer your questions without knowing what you are trying to achieve

cheers Bruce

On Wed, 11 Jan 2017, miracle ozzoude wrote:

Hello freesurfer,

...

I have several questions about mris_expand command. I know it can be used to expand or shrink a point on a surface by a fixed amount or distance. My questions are:

1. Can it be used to expand the pial surface? if yes, please can you

confirm if this is the right script= mris_expand -thickness rh.pial 0.5 rh.newpial. If the script is wrong, please can you give me a directions on how to change it and what new flags to add. 2) When I looked up mris_expand description, -thickness was the only flag displayed. Are there other hidden flags? If yes, where can I find them? 3) After expanding ?lh.white and ?h.pial, what steps of recon-all should i run in other to incorporate my changes? Thank you.

Best, Paul