[Insight-users] problem with ITK ASM

[Deepak Roy]

Hi Louis,

Thanks for the detailed reply. I looked the code in     
 
Insight/Examples/Segmentation/
GeodesicActiveContourShapePriorLevelSetImageFilter.cxx

May be i wasn't detailed enough in explaning my problem
 
I have a set of training images which have ROI (my Shape). Also, I have set of landmark points representing the ROI boundary for each image. 
 
I have calculated a meanshape, eigen values and eigen vectors ie. I have the statistical shape model.
 
I haven't used the first part (ITKASMModelCalculator) to calculate the MeanShape and stats model (because Things were not explained in detail about what is input and output images)

I am trying to use the second part of ITK ASM 
 
I want to do ASM fit on test image and extract ROI by using the ITK Active Shape Model Gradient Search Method 

It looks like it is based on searching for the maximum gradient pixel along a normal line at each landmark point for each iteration and re-adjusts the landmark points ( within the statistical eigen value constraints of the ASM model ).  But this may be prone to noise and not accurate.
 
Can u recommend anything which does ASM fitting very close to implementation in "The use of active shape models for locating structures in medical images" by Cootes et al. paper.
 
Thanks in advance,
Deepak

 

		
---------------------------------
Do you Yahoo!?
 Make Yahoo! your home page   
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://public.kitware.com/pipermail/insight-users/attachments/20050310/dc06108a/attachment.htm