The equation is not quite right. The volume is first computed at each vertex (vertex volume = vertex area * vertex thickness), then it is summed across vertex. This does not lead to the same result as summing the area, computing mean of the thickness, then multiplying them together.
doug
On 10/03/2014 11:42 AM, Wu, Mon-Ju wrote:
I have a quick question about FreeSurfer: is cortical volume the product of cortical thickness and cortical surface are? Conceptually I assume that to be the case, and I found several papers to support that. However, when I tried it with my own data, that seems to disagree with ~12% error.
So I took a deeper look from your website and use the data from your following page as a test: https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/AnatomicalROI_tktools
(#StructName/ SurfArea/ GrayVol/ ThickAvg/ Surface*Thick) (bankssts/ 1133/ 2693/ 2.331/ 2641) (caudalanteriorcingulate/ 805/ 2396/ 2.657/ 2139) (caudalmiddlefrontal/ 2205/ 6275/ 2.528/ 5574) (cuneus/ 1597/ 3129/ 1.806/ 2884) (entorhinal/ 403/ 1837/ 3.268/ 1327)
The above results are took directly from your website and the last column is the product of the surface are and thickness (calculated by me), Theoretically the Gray volume should agree with the product of surface area and the thickness. However, the results above showed about 10.8% error. So I was wondering is my hypothesis of "cortical volume is the product of cortical surface area and cortical thickness" wrong or am I missing something here? What should the relationship be between cortical surface area, cortical thickness, and cortical volume? _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer