Hi,
In our analysis of stroke patients, I've followed the recommendation http://www.mail-archive.com/freesurfer%40nmr.mgh.harvard.edu/msg34040.html to process images, then flip the final surfaces with xhemi for only the subjects with RH lesions to group all healthy hemispheres together for statistical analysis (instead of flipping the initial/raw images prior to processing.)
1 - In documenting the process (and preparing for publication), can you expand upon the reasoning (and possibly point to a reference) for processing prior to flipping? (possibly L-R asymmetries in templates/atlases used in talairach registration, skull stripping, segmentation, parcellation?)
2 - The xhemi reference, Greve et al 2013, explains the symmetric template creation very well. Is fsaverage_sym the template from the ref? (N=42, mostly women)
Thanks, Peggy