[Insight-users] Fast Marching segmentation on 3D image: an empty result volume

[sara meghellati]

Hi,
Sorry to ask again, I looked in the mailing list but I couldn't find an answer to my question.
I'm now testing the level set segmentation code with 3D data. I have started with the fast marching based method. The code works well for 2D image but when I use 3D date (I use Patient1.mhd from http://public.kitware.com/pub/itk/Data/LiverTumor/) I get an empty volume as segmented image. I have checked the output of smoothing, gradient magnitude and the sigmoid and all are fine, just the output of the FastMarchingFilter is empty.
To run this code I have used different seed points (for instance (160,200,6)) but the result still always empty.
I used the same parameter as in page 539 of ItksoftwareGuide for left ventricle and I have used a stopping value of 100.
Regarding the book, this code needs a good speed image to run well and the seed region should be near to the images edges. Perhaps the parameters I'm using are not suitable for this example so could you tell me the appropriate parameters That can be used to make this code running well for this example (Patient1.mhd)? or perhaps there is other thing wrong.
 
Thank you in advance.
Regards
Sara

 
 
 
 


      _____________________________________________________________________________ 
Envoyez avec Yahoo! Mail. Une boite mail plus intelligente http://mail.yahoo.fr
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/pipermail/insight-users/attachments/20080827/9d4d5a2d/attachment.htm>