Hi Jason,
The name of your atlas target spec looks correct. I'm responding from
home, and I don't have the web address for our "Core6 landmarks" web page
handy, but you might try searching the caret-users archives for it. If
you are using the atlas target dataset linked from that page, then you
have no worries. Your trouble is likely visualization-related.
When users first start registering their data, they often make the mistake
of trying to view their registration results (mostly kidding here). By
default, Caret creates a spec file named like
"deformed_my_original_surface_specname.spec" in the atlas target
directory. This spec has a combination of deformed coord files and
whatever coord files were in your atlas target spec (should be only the
atlas sphere). I think it has a deformed topo file, too (not sure), but
this is what David means by you have the wrong topo selected. Not all
coord-topo combinations will look sane when viewed in Caret.
A better way to view your results is to use what we call a visualization
spec. These typically contain the PALS_B12 average fiducial and inflated
surface; closed, open, and cut topo files; and the colin flat map deformed
to PALS_B12. It may also contain the composite paint file and areacolor
file, if you're using partitioning schemes. Then, you can create a
composite surface_shape or metric file containing a single column for each
subject's deformed data (e.g., depth or functional contrast). Once you
have done so, you can switch from one subject's data to the other overlaid
on the flat or inflated surfaces. The result will be more sane.
To create a composite shape, for example, we open the first subject's
deformed shape file; select erase existing columns; set all columns to do
not load; and change the action for deformed depth to Create New Column;
we change the column name from deformed depth to "def subj1 depth" and
press OK. Then we open the second subject's surface_shape file, but this
time we don't erase existing columns. We do select set all columns to do
not load; change the action for depth to create new column; and change the
column name to "def subj2 depth"; and so on for each subject. Save often.
Donna Dierker
> Jason,
>
> I suspect that somewhere along the line, the deformed left hemisphere
> surfaces got linked to an inappropriate topology file. However, the
> atlas spec file you used should be valid - It is a left-right hybrid
> target that is viewed with a right hemisphere topology. We have done
> many Caret registrations from both hemispheres ('left to LR' and
> 'right to LR') to this target.
>
> All I can suggest is that you view the left hemisphere files (pre-
> deformation) in Caret and make sure they all have proper topology and
> hemisphere assignments (e.g., coord files should be designated as
> left hem and should show a lateral view when you click Toolbar: L).
>
> Donna Dierker or John Harwell may be able to troubleshoot in more
> detail based on what you've said. (But Donna is out Monday morning.)
>
> If necessary, we'll ask you to upload the files so we can decipher
> the problem with the data at hand.
>
> David
>
>
> On Apr 14, 2006, at 9:06 PM, Jason D Connolly wrote:
>
>> Dear Donna or David,
>>
>>
>> Thank you very much for your timely assistance thus far.
>>
>>
>> We have successfully morphed a right hemisphere to the B12 atlas. My
>> question concerns the left hemisphere deformation. When I run the
>> spherical deformation, the output left hemisphere deform_ files (flat,
>> inflated, fiducial etc.) have cortex that appears "black" in
>> appearance,
>> very similar in color to the background. Obviously, some error has
>> occurred. Could this have to do with the atlas .spec file that was
>> used
>> for the deformation? I used:
>>
>>
>> Human.PALS_B12.LR.REGISTER-with-INDIVIDUAL.73730.spec
>>
>>
>> as I may have incorrectly assumed the 'LR' refers to left and right
>> and
>> thus the same spec file is to be used for both hemispheres ...
>>
>> ------------------------------------------------------------
>> Jason D. Connolly, PhD
>> Center for Neural Science, New York University
>> 6 Washington Place Room 875, New York, NY 10003
>> cell:646.417.2937 lab:212.998.8347 fax:212.995.4562
>> http://www.psych.nyu.edu/curtislab/people/jasonconnolly.html
>>
>>
>> _______________________________________________
>> 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
>
