No cropping is needed, but the "paint volume" (aka label/ROI/parcellation
volume) needs to be in register, volumetrically, with the anatomical volume
used
to generate the fiducial surface. The voxdims need not match, but the origin
and rotation must match. Headers must correctly identify the orientation in
both the anatomical that generated the surface and the paint volume.
We don't normally segment population anatomicals, but if you managed to do this
successfully, I hold the lecture.
When you try to load the parcellation as paint, it wants an area color file.
Are you loading it as a paint volume? Unless it's an AFNI volume with
Caret-specific header tag like "LUT_Names" or a 4dfp.ifh file with an index to
label lookup, I'm not sure how it will map the integers in the volume to labels
like Brodmann1 or MT. Could you elaborate?
If the parcellation and anatomical that generated the fiducial are
stereotaxically in register (as I'd expect), then your worries in 1) and 3) are
overblown, but you do need to define area colors.
If you are getting messages like "no color for my_label_name", then you can
define colors using Attributes: Area Colors: Edit Area Colors. If there are
many labels, you can create an area color file in a text editor like so:
http://brainvis.wustl.edu/CaretHelpAccount/caret5_help/file_formats/file_formats.html#areaColor
________________________________
From: Long Chuan <[email protected]>
To: "[email protected]" <[email protected]>
Sent: Sun, November 6, 2011 5:00:12 PM
Subject: [caret-users] about mapping the cortical parcellation scheme on
fiducial surface
Hi, Caret experts
I am a Caret newbie and have a few questions regarding mapping a customized
cortical parcellation scheme to a fiducial and/or inflated surfaces in Caret.
My situation is that I created a mask with multiple cortical regions on a
population-specific template using Matlab codes and want to map this
parcellation scheme on a fiducial surface in Caret. So far this is what I have
done:
1) Segment the template and created fiducial and inflated surfaces for the
template. As required by Caret, the brain is cropped, resampled and only left
hemisphere is created.
So, my questions are:
1) What preprocessing steps that I need to map the parcellation scheme on
the anatomical template, since they have different voxel sizes, origin and
coverage (the parcellation scheme still has whole brain coverage and matches
with the anatomical template in FSL).
2) When I try to load the parcellation scheme as paint, I always get the
warning message saying the areal color file is needed to load paint. How can I
get/create this areal color file that corresponds to the parcellation scheme?
3) Based on my search on tutorials, it seems no special preprocessing
steps
are needed for functional/tractography results to be mapped to the fiducial
surfaces. Considering that the tractography results, for example, may have same
problems as the parcellation scheme that I have (different voxel resolution,
origins and brain coverage with the template, after the template is processed
in
Caret) am I missing some documents that introduce the procedures required for
this type of mapping?
Any inputs or suggestions will be greatly appreciated!
Leon
_______________________________________________
caret-users mailing list
[email protected]
http://brainvis.wustl.edu/mailman/listinfo/caret-users
