[Insight-users] 3D segmentation by GeodesicActiveContourShapePriorLevelSetFilter ?

[xiaolong li]

Hello, everyone
   I want to try apply itk::GeodesicActiveContourShapePriorLevelSetFilter on the 3D image segmentation. i have went through the example in tutorial, Here i have some questions, if anyone has done this before, please help, thanks a lot !!
1) what will be the input and output file format for the itkSignedDanielssonDistanceMapImageFilter in 3D case?
 for now i have 10 series of dicom images, each has 40 slices, i will manually segment each slice to binary image(*.png)
   So do i need to combine those 40 images to one mha file as input for the distance map image filter? 
   if yes, how to do that?
   if 10 series of images has different x,y resolution and z thinkness value. how to normalize?
2) what will be the input and output file format for the itkImagePCAShapeModelEstimator in 3D case?
   
3) in the example of GeodesicActiveContourShapePriorLevelSetFilter, there are three seed points manually defined, however,  in general case, especailly in 3D segmentation, is it necessary to do that? or any other alternative way?  
i found very limited info about this topic on internet, i really appreciate if anyone can help to answer those questions, or give me some information related to this. 

btw, thanks luis for answering my previous question, very detail and helpful.
Best Regards
Leonid


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/pipermail/insight-users/attachments/20080526/7acfa31d/attachment.htm>