Hi all,
I was wondering, for source estimates in the fsaverage space, is there a
quick way of determining the mirror vertices bilaterally? (e.g.vertex 0 is
in the left hemisphere, what's the opposite in the right?)
Sorry if it's explained somewhere in the tutorials, hadn't come across.
Many thanks,
Rezvan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20160107/df6c6723/attachment.html
Sorry, my bad, sorted. the location for 0 does pair with 10242, 1 with
10243 and so on... thought had checked this before.
It is important to note: while this may be true for fsaverage (I haven't
done a check of this) it is definitely NOT true for any individual
participant's brain.
hth
d
Yes, absolutely not for single subjects; there the numbering is in the
original space and not in the downsampled space and not equal number of
vertices for the two hemispheres.
For the stc files defined in the downsampled space there seems to be a
rough correspondence.
Is there a way I can accurately find the mirror vertices?
Thanks,
Rezvan
I would recommend another technique for a couple of reasons:
1. The numbering system only has a rough correspondence for many
vertices and can be wildly off for others. (which makes that more difficult)
2. actually calculating a fit that makes sense is extremely difficult
and is exactly what FreeSurfer is designed to do, which is what the
purpose of the morphing options in mne are designed to take advantage of.
So experimenters can morph to an individual (either from your experiment
or fsaverage, or an average from your experiment). Then extract the data
from similar labels in the annotations.
If those don't fit your needs and the experimenter truly wants the
contralateral non-functionally defined region (this is very specific to
the region and reasons that the experimenter is interested in it), then
look into the xhemi work (Doug Greve is the major developer of it) found
here:
http://surfer.nmr.mgh.harvard.edu/fswiki/Xhemi
This page has a link to the paper which should be read.
Note, that there is a separate fsaverage_sym which is different from
fsaverage (this may mean it is not safe to assume the numbering is
symmetrical, though this would be a question for the freesurfer list). I
think if you want true symmetry between the surfaces this is the
technique to apply. Think carefully though before going down this road:
real brains are not symmetrical, so in many cases it will not make sense
to head down that road.
hth
d
Thanks a lot Dan, very helpful suggestions.