[Insight-users] Fast Marching Segmentation

Luca Antiga luca.antiga at gmail.com
Thu Jul 5 15:36:19 EDT 2007


Hi Nisha,
  please let me know if you need suggestions regarding the colliding  
fronts filter
(or about vessel segmentation in general).
It would be useful to have an image showing what your problem looks like
Regards

Luca

--
Luca Antiga, PhD
Head, Medical Imaging Unit,
Bioengineering Department,
Mario Negri Institute
email: antiga at marionegri.it
web: http://villacamozzi.marionegri.it/~luca
mail: Villa Camozzi, 24020, Ranica (BG), Italy
phone: +39 035 4535-381


On Jul 5, 2007, at 3:31 PM, Luis Ibanez wrote:

>
> Hi Nisha,
>
>
> If you are interested in segmenting vascular structures
> you may want to try the itk::CollidingFrontsImageFilter
> http://www.itk.org/Insight/Doxygen/html/ 
> classitk_1_1CollidingFrontsImageFilter.html
>
> that was contributed by Luca Antiga.
>
> ---
>
> Otherwise, in the FastMarching filter itself, the two best
> actions to take in order to improve your segmentation are:
>
>
>    1) Improve the speed input image to have stronger edges
>       between veins and arteries
>
>
>    2) Find a better threshold of the time crossing map
>       that is the output of the FastMarching filter.
>       You may find that a lower threshold value of the
>       time crossing map the vein and arteries "may" be
>       separated.
>
>
> You may also find useful to pre-process the image with
> any of the anisotropic diffusion filters in ITK.
>
> You will find descriptions of these filter in the
> ITK Software Guide
>
>
>   http://www.itk.org/ItkSoftwareGuide.pdf
>
>
>
>    Regards,
>
>
>       Luis
>
>
> ------------------------------------------
> Mukherjee, Nisha (NIH/CC/DRD) [F] wrote:
>> Hello,
>>    I am using the Geodesic Active Contours Segmentation Level Set  
>> to segment the pulmonary veins. However, the fast marching filter  
>> seems to be too inclusive, always segmenting the pulmonary  
>> arteries along with the veins. How might I fine-tune the  
>> parameters of the fast marching segmentation level set to produce  
>> a more precise segmentation? Thank you.
>>  -- Nisha
>> _______________________________________________
>> Insight-users mailing list
>> Insight-users at itk.org
>> http://www.itk.org/mailman/listinfo/insight-users
> _______________________________________________
> Insight-users mailing list
> Insight-users at itk.org
> http://www.itk.org/mailman/listinfo/insight-users



More information about the Insight-users mailing list