i think part of the confusion results from the option in qdec to import the ROI thickness values into qdec for selection as factors in an analysis. if what you want to do is simply look for cortex-wide (ie vertex-wide) differences in thickness between two groups (patients vs. controls), then you dont need to import thickness stats as factors.

if you wanted to do an ROI analysis of patients vs controls, then you wouldnt use qdec, but any old stats package, importing the data from the freesurfer stats tables.

so try running a sraightforward analysis in qdec, with a qdec.table.dat like this for example

fsid group subj1 patient subj2 patient subj3 control subj4 control

and look in the area of banksts for changes if you expect them there. there is a multiple comparison correction option as well.

n.

On Sat, 2011-02-26 at 16:38 +0800, 汪贵宏 wrote:

Hi all,

I want to analyse the thickness difference between patients and normals,but I can not understand the results that I got from qdec-analysis .I hope someone can help me.

Firstly,as the papers says, the Cortical (surface-based) Analysis measures the Distance between white and pial surfaces , get One value per vertex in result.And also,there are 33 areas when we choose to analyses,and we can only get the thickness of these 33 areas per hemisphere,so,my question is why it has just one thickness value in such a large area on the hemisphere?

Secondly,in my opinion,there will be red or green in a large scale if it does exist thickness differences between the patients and normals in one areas,such as "lh_bankssts_thickness",but,the results seems not in that way,why?

Do I explain my questions clearly?

Thanks,

wang

---

Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer