Hi Daniel,
1. Do you mean the atlas or the parcellation of your subjects? We cannot distribute the individual subjects in the atlas as the data was collected under an IRB long ago that didn't include this. The atlas itself (summary statistics derived from those subjects) is included in the FreeSurfer distribution.
2. The subcortical areas are the same, but not the cortical ones. That cortical parcellation is quite coarse, and while we did implement it we were never given permission to distribute the automated version, so it is not part of FreeSurfer. The boundaries that we define are thus different than that one (and since the Harvard-Oxford one is based on volumetric labeling of folding patterns we found that it contained a bunch of errors in which e.g. the wrong side of a sulcus was labeled)
cheers Bruce
On Wed, 3 Oct 2012, Daniel Ferreira wrote:
Dear experts, I have a couple of questions regarding the cortical parcellation atlas that is used in Freesurfer:
- I would like to know if it is possible to obtain the Desikan-Killiany
atlas in 3D analyze/nifti format. With each region of the brain assigned a discrete integer number (for example the all of the voxels of the left inferior temporal lobe have an intensity = 1, or the whole of the left hippocampus = 2).
- Are the Harvard-Oxford cortical and subcortical atlases that can be
downloaded with FSL the same as the cortical parcellation atlas that is used in Freesurfer? (i.e. are the neuroanatomical boundaries the same; does the left inferior temporal cortex in the Freesurfer cortical parcellation = the left inferior temporal cortex in the Harvard-Oxford atlas downloaded with FSL)
Many thanks
Daniel Ferreira