There is not such a script. You can do a different check though by computing the mean area on the crown and in the fundus in the native subject space, then comparing these numbers across subjects to see if the effect holds. This test gets around the problems of sampling areal measures into a standard space. To do this, first load a subject in tksurfer with the curvature as an overlay
tksurfer subject lh inflated -overlay lh.curv
Play with the overlay threshold until you get something that looks about right. Then run mri_segstats like
mri_segstats --i lh.area --mask lh.curv --mask-thresh yourthreshold --annot subject lh aparc --o subject.summary.dat
You can then use asegstats2table to combine all the subject.summary.dat files into a single table, then use mri_glmfit passing it the final table with --table to do the final stats (look at the text file output).
doug
On 03/26/2014 07:36 PM, Paul Mazaika wrote:
We are using Freesurfer v5.3 with the recent Jacobian area correction, but there may be a subtle bug that affects longitudinal studies. In a pediatric population, we see area growth on top of gyrii, and area loss in the base of sulci. This effect may be a true morphological change, but it may be due to an approximation error of a curved area by planar surfaces in regions of high curvature. Is there a command or script to compute an "adjusted area" on a vertex-wise basis? For every vertex, I would like to compute something like: Adjusted area = Area*(1 + G*d^2*k^2) where Area is the v5.3 area, G is a constant, d is the local vertex spacing, and k is the local mean curvature.
Thank you, Paul K. Mazaika _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer