[Insight-users] liver vessel segmentation

[Luis Ibanez]

Hi Rodolfo,

If the blood vessels are at least 5 pixels in diameter in
your images you could use the CURVESLevelSetImageFilter.

http://www.itk.org/Insight/Doxygen/html/classitk_1_1NarrowBandCurvesLevelSetImageFilter.html


Implemented from:

   L. Lorigo, O. Faugeras, W.E.L. Grimson, R. Keriven, R. Kikinis,
   A. Nabavi, and C.-F. Westin,

   "Curves: Curve evolution for vessel segmentation."
    Medical Image Analysis, 5:195-206, 2001



Otherwise you may consider computing the Hessian of the image and
tracking along the direction of the minimal Hessian eigen-value.



As in


    "Ridges in Image and Data Analysis"
    by David Eberly


and


   "Initialization, noise, singularities, and scale in height ridge
     traversal for tubular object centerline extraction"

    Aylward, S.R.; Bullitt, E.;

    Medical Imaging, IEEE Transactions on  ,
    Volume: 21 , Issue: 2 , Feb. 2002
    Pages:61 - 75



This is illustrated in 2D in the ITK Application:


    http://www.itk.org/HTML/Curve2DExtraction.htm






    Regards,



        Luis


------------------------
Rodolfo Palacios wrote:

> Hi,
> 
> I'm working on the segmentation of liver vessels (arteries, veins and 
> biliary conducts). The input images are in Dicom format obtained from CT 
> Scans and I'm using VTK for visualization. My question is: Which one of 
> ITK's segmentation methods (hybrid methods maybe?) is better suited for 
> this kind of job considering that the contrast between vessels and liver 
> tissue is very small?
> 
> Thank you.
> 
> Rodolfo Palacios.
> 
> 
> _______________________________________________
> Insight-users mailing list
> Insight-users at itk.org
> http://www.itk.org/mailman/listinfo/insight-users
> 
>