Thanks for the info Doug. My comments below.
On Mar 10, 2004, at 1:19 PM, Doug Greve wrote:
Ray Fix wrote:
Hello List,
Question 1: Where does the sample average7 come from? I searched through the manuals, googled, searched around a little but was unable to find any info. For example, is this the averaged brains of 7 male subjects in there 40s?
It's a 7th order icosahedron (163842 vertices). I think the orig coordinates are supposed to be talariach coordinates averaged over some subjects (but it does not overlay very well on the talairach brain). Bruce can give you more details.
Wow. Interesting. I never would have guessed. Thanks.
(Aside: Let's see ... 20 * 7^2 is only 980 faces so I am not sure where all of those vertices are coming from. Probably some different definition of order 7.)
Question 2: My plan is to create my own ICBM labels using average7 and then apply them via mri_label2label to two groups of subjects (which happen to be pediatric data). Does this stand any chance of working? I would be very interested if their are any other approaches you recommend.
It all depends on how you're going to map the labels. What coordinate system are the ICBM labels in?
I guess I have to look into this more. Any suggestion on what coordinate system should I use? Perhaps talairach. To be honest my only knowledge about labels comes from looking at the recon-all script and the one Faq-O-Matic page.
I was told that the labels in FreeSurfer are sort of non-standard. The whole motivation for reporting results in terms of ICBM labels is to make the results meaningful a large audience without getting into a huge discussion about label definitions.
That said, it is likely that deep consideration was given in coming up with the current default labeling scheme. Maybe someone would like to refute my assertion that they are non-standard. It is something I would like to learn more about but am not sure how to find out.
Thank you very much for your comments.
Ray Fix