To clarify, F6 is a six-subject macaque fascicularis atlas developed by
a collaboration of wustl.edu researchers. See part 3 of the tutorial
cited below for details.
On 09/20/2006 08:12 PM, David Vanessen wrote:
John (et al.),
I guess Donna wins a six-pack, because I do indeed have a response.
1) First, some general comments. Your questions come at a propitious
time.
We are just completing a major revamping of the macaque F99 atlas
along with many other changes to the macaque and human (PALS) atlas
datasets, Caret 5.5 software, and a tutorial that illustrates many of
the new features. Our intent is to announce this to the Caret user
community in the coming week. (We just gave a successful 1-day
workshop at Wash U and are now cleaning up on various details that
emerged from this session.)
More specifically on the macaque atlas front, what we are calling the
new 'F99' atlas differs from the old 'F99UA1' atlas in several ways.
Most important is that both the left and right hemispheres are in a
'standard-mesh' format (with 73730 nodes, same as our human PALS
atlas). This greatly facilitates simultaneous visualization of the
macaque left and right hemispheres (as well as concurrent
macaque-human visualization) in a single Caret application.
Another aspect of this is that we registered the old F99UA1 left and
right hemispheres to a 'hybrid' left-right target, in which the
landmarks were generated by averaging the right and (mirror-flipped)
left hemispheres, so it is not biased by one side or the other. Also,
the number of landmarks used for registration to the older F99UA1
surfaces is 26, larger than used previously. The number used for
registering to F6 is smaller (24) and to the Paxinos (PHT) atlas is 23
- i.e., they are case-specific, depending on what is reliably
distinguishable in the source and target.
2) More specific to your starting question, the starting border
projection file for the new F99 atlas is
http://sumsdb.wustl.edu/sums/archivelist.do?archive_id=6590483&archive_name=Macaque.F99UA1.LR.AVERAGE.LANDMARKS_Surfs-29407_29830_for_FULL-HEMISPHERE.73730.borderproj
An archive containing all three LANDMARK borderproj files has just
been uploaded to SumsDB and is available at:
http://sumsdb.wustl.edu/sums/directory.do?id=679531&dir_name=MACAQUE
http://sumsdb.wustl.edu/sums/archivelist.do?archive_id=6589983&archive_name=Macaque.F99UA1.REGISTER-with-INDIVIDUAL.73730.spec
If you view your individual hemisphere and the F99 atlas surface in
separate Caret applications, determine how many of the 26 F99
landmarks are reliably visible in your individual hemisphere. If some
landmarks are unreliably (e.g., the posterior spur of the arcuate
sulcus = ASp), delete it from the F99 borderproj file and save that by
a different name. If other landmarks are visible in both hemispheres,
but you judge the extent where there is 'correspondence' to be
different, then you can 'nibble' the F99 landmarks (delete border
points) or re-draw the landmark border, then reproject and re-save the
modified F99 version while generating a corresponding landmark in the
individual.
Before you attempt spherical registration, make sure that the number
of borders in the individual and atlas borderprojection files are the
same. Use the query (?) button in the open borderproj file or in the
spec file, and select Border Names, which gives the total number as
well as the individual names.
The datasets and the atlas tutorial are in
CARET_TUTORIAL_SEPT-06
http://sumsdb.wustl.edu/sums/directory.do?id=6585200
We are aiming for an update of the tutorial and datasets to be
inserted/released by the weekend.
I hope this helps....
David VE
On Sep 19, 2006, at 2:58 PM, John Arsenault wrote:
Hello,
I was wondering which .borderproj file was appropriate for the
registration of an individual macaque left hemisphere to the left
hemisphere of the F99 ATLAS. I found a file named
"Macaque.F99UA1.LANDMARKS_for_RegWithMacaque.R.03-04-21.34132.borderproj"
in the left hemisphere section of the macaque atlas on SuMSDB. Is this
the correct file to use for spherical deformation? This .borderproj file
has more borders then there were for the .borderproj file I used for my
right hemisphere sperical deformation with the same monkey
("Macaque.F99UA1.R.LANDMARKS_for_FREESURFER-REGISTRATION.03-05.35946
.borderproj") incuding a posterior version of the AS border and
(LANDMARK.CaSv, LANDMARK.MW_HFtoTempCut, LANDMARK.PostOrb, LANDMARK.LOS,
LANDMARK.VLC, LANDMARK.DMC, LANDMARK.VMC, LANDMARK.OrbS,
LANDMARK.MW_POStoCaS). If I should use these borders then what
deformation map file should I use? If this is not the file I should use
then which file is correct for left hemisphere deformation? Also with
the
updates to f99 what borders and .borderproj files will be used for
registration with the updated Atlas and when will that be available.
Thank you so much for your help,
John
_______________________________________________
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
--
Donna L. Dierker
(Formerly Donna Hanlon; no change in marital status -- see
http://home.att.net/~donna.hanlon for details.)
