Hi,
As explained in the paper, the radius of the sphere correspond to the region of interest in which you analyze gyrification: if you take 25mm, you will measure all the amount of cortical surface enclosed in a 25 mm radius from each of the cortical point (that is the default). You can chose whatever you want, keeping in mind that reducing the radius will give you a better regional discrimination and that increasing it will smooth your results. In decreasing the radius, the limiting factor is that you want to measure the effects of many sulci on gyrification, so that I would not chose below 15 or 20 mm. In the paper you have a graphic with the smoothing of the distribution with increasing radius. At the end, (if you stay within reasonable limit, e.g. 15-50mm) it will not change much the statistical significance of your results, but it will change their regional accuracy. Is that clear? You can always try with a limited set of data, and you will quickly understand of lGI behaves according to varying radii.
Marie
On 27 oct. 09, at 15:08, Zhangyuanchao wrote:
Hi, expert,
Could anyone tell me the effect of the radius on lGI. After reading the paper on lGI carefully, I still can not get a clear view of the effect of the radius on lGI. I still wonder why the default radius are chosen to be 25mm since there is very little discussion of the standard in choosing such a radius in the paper on TMI. When making group analysis, does the result vary greatly using different radii? and why is that? How to intepret the result obtained using different radii?
Thank a lot!
Best Regards
好玩贺卡等你发，邮箱贺卡全新上线！ _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer