Hi all,
I'm curious about how to select vertices based on their overall location.
For example, say I have found two peaks in my source estimates. I want to
build two functional labels for each. One blob is in the anterior half and
the other is in the posterior half of the left hemisphere. Are the vertices
organized in such away that I can select each blob on the basis of its
overall location in the brain?
Thanks in advance!
You can do this in mne_analyze by simply drawing around the ROI, and
selecting (the procedure is described in the manual). You can also do
this programmatically in the mne-python code (this is a good starting
point: http://martinos.org/mne/stable/auto_examples/inverse/plot_label_from_stc.html?highlight=label
). or with a simple bit of self-written matlab/python code depending
on your needs.
hth
d
I have been working off of that example that you linked in. The
mne.stc_to_label is very useful and gets me part of the way there but I
still can't say which is anterior and which is posterior until I plot them.
Which is fine with two labels but could get complicated with more (or to
iterate across many subjects). If I can avoid the manual option, that'd be
great.
Really all I'm looking for is access to the spatial location of each vertex
in some kind of xyz coordinate space. With that, the rest is, as you
mentioned, a bit of simple self-written code but I haven't been able to
find any documentation on how the vertices are sorted or structured.
Thanks!
Keith
Hi Keith,
The xyz coordinate for each vertex is in the source space. you can
just calculate anterior vs. posterior using the locations there.
hth
d
Thanks for bearing with me.
Based on shape, It seems as though they are in either the 'nn' or 'rr' keys
of the source space dict. Maybe this is obvious, but what is the difference
between these?
After that I can use vertno to select the correct vertices?
Keith
rr is what you want.
In the pdf manual you would find the description in the matlab toolbox
chapter (there is a description of all the fields). I haven't been
able to find these things after the C-manual was removed from the wiki
(it is probably there, but I don't know the new layout).
hth
d
look at that! There is a manual!
Thanks for your help!
Keith