On 1/25/19 7:52 AM, Klaas Bahnsen wrote:
External Email - Use Caution
Dear Freesurfer Experts,
First of all: Thank you so much for your work and your fantastic product.
I turn to you with a short question, I hope it is not too unnecessary and can be answered quickly. My interest lies in vertex wise longitudinal investigation of cortex surface area. The paper 'False positive rates in surface-based anatomical analysis' (Greve, Fischl 2018) briefly describes how Freesurfer calculates cortex surface areas: First the transfer to the target atlas with "Jacobian Correction" and second the division by eTIV (ICV) to include the factor of different head sizes. Finally, a smoothing should be applied. My question is where these data are stored in the Freesurfer data structure or where they are generated. My understanding is that first recon-all stores this data per scan under SubjectID/surf/?h.area. These could then be picked up by mris_preproc and smoothed with mri_surf2surf. This smoothed file could then be used for example for a GLM model (mri_glmfit). Are the steps described in the paper covered with this workflow? Or must a special option be selected at some point, e.g. to perform the division by the eTIV.
You must divide by the eTIV yourself (though it would be a good feature to add to mris_preproc). Note that mris_preproc will do the jacobian correction.
I'm a bit unsure at the moment because I also found a collection of questions from previous courses on the FreeSurfer Wiki page. In course 2014 Copenhagen the question is asked which units are used in the output files which are calculated by mris_preproc. Part of the answer states that the data in ?h.area.mgh on vertex level do not have much meaning, because they are dependent on the triangle mesh.
Hopefully you can help me and hopefully I won't steal you too much time with my question.
With kind regards, Klaas Bahnsen
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer