[Insight-users] help about itk::ImagePCAShapeModelEstimator and itk::GeodesicActiveContourShapePriorLevelSetImageFilter

[Baoyun Li]

Dear Luis and All:

I am planing to apply some statistical model for CT liver segmenation. After some researches, I think I can  use itk::ImagePCAShapeModelEstimator  to get the average shape and feed the average shape to itk::GeodesicActiveContourShapePriorLevelSetImageFilter. But I am still confused regarding the how to get the shape in commen space( maybe that is the registration problem)

1 I have ten segmented livers from CT image, if the liver masks are in the same space, I understand I can get the signed distance map and then feed them to itk::ImagePCAShapeModelEstimator   to get the average shape.
2 But how can I put the those binary mask in the same space? Should I register the grey level liver image or the binary mask to somewhere?
3 After I got the average shape, I also need to make the average shape and the new CT data in the same space? Can you please suggest me a way to do it? ( My feeling is that registering the different CT abodomi data is quite difficult. I tried to use rigid and affine registration tool provided by Daniel Rucker, it is hardly to see any registration effect for CT abodimi images of different person).


Thanks and my best regards

Baoyun


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/pipermail/insight-users/attachments/20090527/5d840ff4/attachment-0001.htm>