Print

Print
Hi Steve,
 
The simple averaging scheme did not yield good results. Looks like I have to go the hard way and work with mesh representations to find the average shape from the segmentations.
 
The FIRST website says that
"The manual labels are parameterized as surface meshes and modelled as a point distribution model. Deformable surfaces are used to automatically parameterize the volumetric labels in terms of meshes; the deformable surfaces are constrained to preserve vertex correspondence across the training data." 
 
This is exactly what I want. I have a set of volumetric labels which I want to parameterize and find the average volumetric label. How can I go about this  ? Also I hear there is going to be a new patch for FIRST. Are there any plans for new tools to be updated to FIRST_UTILS which can perform this function?

Thanks for your help.
 
Thanks,
Jayanth.

On Thu, Mar 12, 2009 at 5:08 AM, Steve Smith <[log in to unmask]> wrote:
Hi,


On 11 Mar 2009, at 18:37, Jayanth Kolar wrote:

Hi All,

I have a set of 3D binary masks of segmented ventricles and I would like to generate an average mask from the set of 4 masks.Basically I want to create parametric mesh surfaces from the segmented ventricle masks and average corresponding points.

I believe that of the two possible routes here one is easy and one hard - it would be easy to take the 3D image volumes of the masks, transform them into a standard space (if they're not already) and average-then-re-threshold using fslmaths. The hard way is to work with the mesh representations instead, for which I think you would probably need to get inside the source code to do this kind of thing - Brian/MJ can correct me if I'm wrong.

Cheers, Steve.




Is there a way I can do it for such arbitrary labeled volumes and obtain point correspondences using FSL-FIRST or any other tool ?

Thanks for your time and thanks for the great tool. We have started using it for almost everything !

Regards,
Jayanth.

--
Jayanth Kolar
Graduate Research Assistant
Physiological Imaging and Modeling Lab
University of Illinois at Chicago
Chicago IL - 60607
1-408-398-0097
http://www.ece.uic.edu/~jkolar


---------------------------------------------------------------------------
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director,  Oxford University FMRIB Centre

FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK
+44 (0) 1865 222726  (fax 222717)
[log in to unmask]    http://www.fmrib.ox.ac.uk/~steve
---------------------------------------------------------------------------



--
Jayanth Kolar
Graduate Research Assistant
Physiological Imaging and Modeling Lab
University of Illinois at Chicago
Chicago IL - 60607
1-408-398-0097
http://www.ece.uic.edu/~jkolar