 |
 |
Hallo Svenja, I'm developing a process chain to perform a similar task on tractograms. I think it can be used with fMRI too. 1) map your blob to the histological atlas -compute linear and nonlinear map from source images to atlas space - apply transformation to your blob 2)mask the blob with all labelled region on the atlas and compute the sum. 3) divide by the blob and mult. 100 You have percentages. Pay attention that the results are so precise as your mapping to the atlas. So if the brain variability is high and the mapping is poor this method is clearly suboptimal. Practically you can apply it to broad regions where the misfit is small compared with the size of the blob. Second possibility consists in marking the regions manually for each brain and use manual segmentation instead of Juelich atlas. Best. Davide. On Fri, 2009-03-13 at 11:20 +0100, Svenja Diekhoff wrote: > Hi there, > > I was wondering whether it is possible to get information about how > many percent of a certain fMRI activation blob are located within a > certain brain area using the Juelich Histological Atlas, e.g. 52 % > BA4a, 30% BA4p, 18% BA6 or something like that. > > I would be very happy about any comments or ideas! > > Kind regards, > Svenja