Re: [caret-users] Flattening and Spherical Registration of the Monkey Brain

2007-02-26 Thread Donna Dierker

On 02/23/2007 02:05 PM, Reza Rajimehr wrote:

Dear Donna and David,

Thank you for the helpful comments and suggestions. I successfully
finished flattening of an individual human brain and its spherical
registration to the human PALS atlas, and the registration is perfect. Now
I want to start flattening of an individual monkey brain and its spherical
registration to the macaque F99UA1 atlas. Before starting this process, I
just wanted to make sure that I am using the correct (and the most
up-to-date) files. In particular, I would appreciate if you provide me
with the SumsDB links for the following files:

1) Macaque template cuts for flattening (border file and border color file).
  
These files come with the Caret distribution and should load 
automagically when you select Macaque standard cuts left or right:


caret/data_files/flatten_borders/Macaque.TEMPLATE-CUTS.LEFT-HEM.SPHERE.border
caret/data_files/flatten_borders/Macaque.TEMPLATE-CUTS.RIGHT-HEM.SPHERE.border
caret/data_files/flatten_borders/Macaque.TEMPLATE-CUTS.STANDARD.bordercolor

2) Macaque landmarks for registration (border color file and border
projection file).
  
Going to the SumsDB home page (http://sumsdb.wustl.edu/sums/index.jsp) 
and clicking the Macaque left Go to Data directory link, then EXTENDED 
DATA SETS, I see this spec:


Macaque.F99UA1.L.ForREG-with-F99UA1.R.03-03.34132.spec
http://sumsdb.wustl.edu/sums/archivelist.do?archive_id=6094474

The dataset for the right is less clear; I'll need to do some digging.  
This might be it:


Macaque.F99UA1.R.LOBAR.35946.spec
http://sumsdb.wustl.edu/sums/archivelist.do?archive_id=6094366


3) F99UA1 atlas for registration (.spec, .deform_map, and sphere.coord
files).
  
See above.  You need to draw the borders for the source (individual) 
macaque, but the atlas target exists somewhere in sumsdb.

4) The complete F99UA1 atlas containing various .coord files (and other
Caret files) for surface visualization.
  
Try the standard scenes and other macaque specs from the Sep 2006 
tutorial (part 3):


CARET_TUTORIAL_SEPT-06
http://sumsdb.wustl.edu/sums/directory.do?id=6585200



and a brief question about flattening: where is the location of temporal
and frontal cuts on the macaque cortex, and how can I draw them? (i.e.
where are the recommended start and end points for these cuts?)
  
I'm not sure.  Unless they're in the landmark border set used during 
registration, it doesn't matter that much.  Not all flattening cuts are 
used during registration (e.g., only calcarine and medial wall are used 
in humans).  In monkeys, there are multiple landmark sets out there, 
with varying numbers of borders.


If they just affect flattening -- not registration -- then drawing a cut 
from the medial wall to the temporal pole will reduce distortion.  For 
the frontal cut, try drawing one from the medial wall along the medial 
orbital gyrus to the frontal pole.

Thanks,
Reza


  

Hi Reza,

The atlas target dataset you used was the right one, but it is a
stripped down version that minimizes the number of files that get
deformed from the atlas to the individual's directory.  (Some people
want to look at the atlas paint or border files on their individual's
surface, so then you would need to put those files in the atlas target
directory and add them to your atlas spec file; then, they would
automatically deform to the individual's directory during registration.)

Since you want to visualize on the atlas target, there's no need to
re-register; rather, just download one or more of these datasets:

PALS_B12.LEFT.STANDARD-SCENES.73730.spec
http://sumsdb.wustl.edu/sums/archivelist.do?archive_id=6595803

PALS_B12.RIGHT.STANDARD-SCENES.73730.spec
http://sumsdb.wustl.edu/sums/archivelist.do?archive_id=6595827

There are quite a few other options/documents listed here:

CARET_TUTORIAL_SEPT-06
http://sumsdb.wustl.edu/sums/directory.do?id=6585200

You can move your deformed results to the extracted directory, or move
the extracted files to your atlas target directory -- your choice.  I
tend to create separate right and left subdirectories under my atlas
target directory, which has lots of files, but when I run registration I
use the single parent atlas target directory.

Often, we want to do group analysis of functional (metric) or anatomical
(surface_shape) data; in these cases, we create composite metric or
depth files.  This command line utility (caret_command -help-full) can
be helpful:

  caret_command -metric-or-shape-composite
  caret_command -metric-or-shape-composite-named-column

If you're going to use the tools under Attributes: Metric and Surface
Shape Statistical Operations, then the spec files above lack the
distortion metrics and OPEN topo files included in these analysis
datasets:

* Visualization/Analysis spec LEFT:
http://sumsdb.wustl.edu/sums/archivelist.do?archive_id=6486316
* Visualization/Analysis spec RIGHT:
http://sumsdb.wustl.edu/sums/archivelist.do?archive_id=6486393

Most of the 

Re: [caret-users] Flattening and Spherical Registration of the Monkey Brain

2007-02-23 Thread Reza Rajimehr
Dear Donna and David,

Thank you for the helpful comments and suggestions. I successfully
finished flattening of an individual human brain and its spherical
registration to the human PALS atlas, and the registration is perfect. Now
I want to start flattening of an individual monkey brain and its spherical
registration to the macaque F99UA1 atlas. Before starting this process, I
just wanted to make sure that I am using the correct (and the most
up-to-date) files. In particular, I would appreciate if you provide me
with the SumsDB links for the following files:

1) Macaque template cuts for flattening (border file and border color file).

2) Macaque landmarks for registration (border color file and border
projection file).

3) F99UA1 atlas for registration (.spec, .deform_map, and sphere.coord
files).

4) The complete F99UA1 atlas containing various .coord files (and other
Caret files) for surface visualization.


and a brief question about flattening: where is the location of temporal
and frontal cuts on the macaque cortex, and how can I draw them? (i.e.
where are the recommended start and end points for these cuts?)

Thanks,
Reza


 Hi Reza,

 The atlas target dataset you used was the right one, but it is a
 stripped down version that minimizes the number of files that get
 deformed from the atlas to the individual's directory.  (Some people
 want to look at the atlas paint or border files on their individual's
 surface, so then you would need to put those files in the atlas target
 directory and add them to your atlas spec file; then, they would
 automatically deform to the individual's directory during registration.)

 Since you want to visualize on the atlas target, there's no need to
 re-register; rather, just download one or more of these datasets:

 PALS_B12.LEFT.STANDARD-SCENES.73730.spec
 http://sumsdb.wustl.edu/sums/archivelist.do?archive_id=6595803

 PALS_B12.RIGHT.STANDARD-SCENES.73730.spec
 http://sumsdb.wustl.edu/sums/archivelist.do?archive_id=6595827

 There are quite a few other options/documents listed here:

 CARET_TUTORIAL_SEPT-06
 http://sumsdb.wustl.edu/sums/directory.do?id=6585200

 You can move your deformed results to the extracted directory, or move
 the extracted files to your atlas target directory -- your choice.  I
 tend to create separate right and left subdirectories under my atlas
 target directory, which has lots of files, but when I run registration I
 use the single parent atlas target directory.

 Often, we want to do group analysis of functional (metric) or anatomical
 (surface_shape) data; in these cases, we create composite metric or
 depth files.  This command line utility (caret_command -help-full) can
 be helpful:

   caret_command -metric-or-shape-composite
   caret_command -metric-or-shape-composite-named-column

 If you're going to use the tools under Attributes: Metric and Surface
 Shape Statistical Operations, then the spec files above lack the
 distortion metrics and OPEN topo files included in these analysis
 datasets:

 * Visualization/Analysis spec LEFT:
 http://sumsdb.wustl.edu/sums/archivelist.do?archive_id=6486316
 * Visualization/Analysis spec RIGHT:
 http://sumsdb.wustl.edu/sums/archivelist.do?archive_id=6486393

 Most of the tests can be run using the caret_command utility.

 Donna

 On 02/21/2007 01:25 AM, Reza Rajimehr wrote:
 Dear caret-users,

 I did the spherical registration of an individual human brain to the
 human
 PALS atlas. I used the following dataset as the target atlas brain:

 http://sumsdb.wustl.edu:8081/sums/archivelist.do?archive_id=6057499

 A new deformed .spec file has been generated. However, it has only one
 new
 surface, which is a deformed sphere. I would like to visualize the
 functional activities of the individual brain on the atlas brain (e.g.
 on
 the atlas flat map). Should I download a more elaborate PALS dataset and
 add them to the current directory? If so, should I redo the spherical
 registration, or the new PALS surfaces (e.g. inflated, flattened, ...)
 will be automatically linked to the deformed spherical surface that I
 already have?

 Thanks,
 Reza

 
 Reza Rajimehr, MD

 NMR Athinoula A. Martinos Center
 Department of Radiology
 Massachusetts General Hospital (MGH)
 Building 149, 13th St.
 Charlestown, MA 02129

 Email: [EMAIL PROTECTED]
 
 ___
 caret-users mailing list
 caret-users@brainvis.wustl.edu
 http://pulvinar.wustl.edu/mailman/listinfo/caret-users



Reza Rajimehr, MD

NMR Athinoula A. Martinos Center
Department of Radiology
Massachusetts General Hospital (MGH)
Building 149, 13th St.
Charlestown, MA 02129

Email: [EMAIL PROTECTED]