Hi Ed,
If this proves difficult, then upload your fiducial coord and topo file
here:
http://pulvinar.wustl.edu/cgi-bin/upload.cgi
I have an idea I'd like to try, where I project the surface to a volume
(Attributes: Paint: Convert paint to volume) and then use the Volume:
Segmentation: Edit voxels to insert a slab along the cut-face. Then,
the closed interior of the surface can be filled (Fill cavities) and a
surface generated from the resulting segmentation. Then you can use the
Volume: SureFitOperations to generate the curvature and paint files,
specifying where the cut face is. This way, SureFit will do a better
job of identifying the cut face, and you can use spherical registration
rather than flat registration, assuming you want to register your
surface to our atlas (which you may not need to do).
Donna
On 11/10/2005 09:07 PM, David Van Essen wrote:
Ed,
Donna's suggestion is in the right direction, but there may be a
problem that entails a bit of extra 'neurosurgery'.
When you use border drawing to reassign paint column data on a 3D
surface, the reassignment currently occurs not only on the nearest
nodes but also on distant nodes that are occluded from view but lie
directly behind the enclosed border. Hence, when CUT.FACE nodes are
deleted automatically during the flattening process, you may end up
with an unintended hole in a region that you actually want to preserve.
One possible work-around is if you can use the Surface:ROI menu to
select a group of nodes that are exclusively on the cut face (e.g.,
because they have distinct identities in one of your paint columns).
If so, you can select, then reassign them to the CUT.FACE identity in
the Geography column.
If that doesn't work, you can go back to Donna's suggestion, but
follow it up by rotating the surface so that you can view the distal
(previously occluded) CUT.FACE nodes on their own. Then draw a border
around these extraneous regions and reassign them to some other
identity besides CUT.FACE.
These are inelegant options, but hopefully one of them will work. Or
perhaps John has a better suggestion for what to do.
David VE
On Nov 10, 2005, at 4:22 PM, Donna Hanlon wrote:
Hi Ed,
This is a bit tough, because you didn't generate your surface from a
segmentation using the SureFit segmentation feature, but I think
there's a work-around. I think if you find a patch of nodes along
the cut-face and use the Draw Borders feature to assign the paint
name CUT.FACE (using the paint column name Geography) to it, and save
the paint file, then the flattening routine might be happy. Try to
minimize the number of nodes and make sure they're right on the edge
of the cut face.
Donna
On 11/10/2005 02:22 PM, Edward Craft wrote:
Hi,
I'm trying to flatten a portion of macaque visual cortex that I
reconstructed from histological sections using Caret5.3, and I ran
into a
similar problem as Rishi. After cleaning-up and cutting my surface, I
selected Surface | Flatten Full or Partial Hemisphere, Partial
Hemisphere
(Fiducial), and received the following error message: "There is no
geography
paint column which is needed for flattening the partial hemisphere. The
geography paint column contains the 'CUT.FACE' nodes which are
needed to
properly orient the partial hemisphere for flattening".
It may be relevant that when I tried using Surface | Cuts | Draw
Cuts to cut
my surface as described in Tutorial 9, Caret retained only three or
four
distantly-spaced points and gave a rough, straight-line
approximation of the
cut I drew. To get better precision, I ended up cutting my surface by
manually deleting links, instead.
I was hoping you could tell me how to create the necessary cut face
information to flatten my surface, and also what might have gone
wrong with
the Surface Cuts feature (i.e. why I couldn't get it to follow a
irregularly
shaped cuts as in the tutorial). I am uploading my files, in case
they will
help. Thanks,
-- Ed Craft
_______________________________________________
caret-users mailing list
caret-users@brainvis.wustl.edu
http://pulvinar.wustl.edu/mailman/listinfo/caret-users
_______________________________________________
caret-users mailing list
caret-users@brainvis.wustl.edu
http://pulvinar.wustl.edu/mailman/listinfo/caret-users
_______________________________________________
caret-users mailing list
caret-users@brainvis.wustl.edu
http://pulvinar.wustl.edu/mailman/listinfo/caret-users
