[Insight-users] itkDeformableSimplex3DFilter: different spacing of image data

[M.M.Nillesen at cukz.umcn.nl]

Hello,

I'm using the itkDeformableSimplex3DFilter.
It is clear from the documentation that this filter currently assumes that the spacing of the input image is 1. The image data I use, do not have equal resolution in X, Y, and Z direction. In order to be able to use the DeformableSimplexMesh algorithm I'm currently resampling the image data such that the spacing is 1 in all directions. However, resampling of the data results (in my case) in large datasets, containing redundant information. The resampling therefore causes memory problems and is not a very elegant solution. 
It would favorable to compensate for the different resolutions in the itkDeformableSimplex3DFilter-algorithm itself and therefore eliminating the need for resampling of the image data.

Has anyone encountered similar problems and solved this problem by taking into account different spacings of the image data when deforming a mesh? Or is there another itk-class that does take into account different spacings when deforming objects?
I know that all the gradientFilters (for example GradientMagnitudeImageFilter, GradientMagnitudeRecursiveGaussianImageFilter) do take into account different spacings, so computation of gradient information in image data with different X,Y,Z spacing is not problem.

Thanks in advance for your help!

Kind regards, 

Maartje M. Nillesen

Radboud University Nijmegen Medical Centre
833 Clinical Physics Laboratory, Dep. of Pediatrics
Nijmegen, The Netherlands
T +31 24 3668967




Kamer van Koophandel - handelsregister 41055629  / Netherlands Chamber of Commerce - trade register 41055629

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://public.kitware.com/pipermail/insight-users/attachments/20070611/ea4600cb/attachment.html