Hi Martin,
Sorry for the delay.
As for the defaults: the main (and almost only) parameter is the size of the circular region of interest, which you can change if you want. There is a discussion about the rationale for choosing a radius between 20 and 25 mm in the IEEE TMI validation paper (which you will find on the wiki). The main purpose of the LGI was to be sensitive to changes on the level of a few sulci and gyri (which was chosen given the variability in sulcal configuration). If you suspect small gyral changes, you can try to set the radius at 10-15mm or so, I will be interested to know the results you can have in your cohort with small radii.
We did not try in neonatal brains (our youngest patients are 6, for which it is perfectly fine to use "adult settings"). The first obstacle that I would imagine is the accuracy of cortical reconstructions in such brains. Then I would imagine that you need a smaller radius for the circular region of interest in neonatal than in adult cortex, and also that you would need to try a few radii before finding the optimal one... Again, I would be happy to know your observations on that!
Marie
On 24 sept. 08, at 20:15, Martin Kavec wrote:
Hi,
I wonder how are the default FS settings to calculate LGI finetuned? More towards large-scale changes, such as in neonal>pediatric cortex, or would it be possible to study rather focal changes, say on a sulcus/gyrus level? Which parameters should I change to become sensitive to "smaller" LGI changes?
Thanks a lot for input,
Martin _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer