[Insight-users] multiple material marching cubes

[Prashanth]

Bill,
  Thanks for the email. I will look into ITK and segmenting and labeling 
images. And just for the record (and archives) the authors from the 
paper do have a software (http://ccni.wpi.edu/miva.html). They have 
their own file format for the multiple material marching cubes 
application and the documentation on that is limited.
Thanks again
Prashanth

Bill Lorensen wrote:
> OK, now I recall this paper. It works with segmented data, that is
> data that has discrete labels. This will not work on your data. If you
> segment your data with techniques available in software systems such
> as itk (http://www.itk.org), then you can use vtkDIscreteMarchingCubes
> to generate surface similar to those found in the referenced paper.
> Perhaps the authors have software you can try and compare the results
> with vtkDiscreteMarchintgCUbes followed by non-manifold smoothing in
> vtk.
>
> In this talk: http://www.bioontology.org/wiki/images/6/64/BeyondPixels.ppt
> you can see some results of vtkDiscreteMarchingCubes run on the
> Visible Human data.
>
> Bill
>
> On Tue, Apr 8, 2008 at 4:37 PM, Prashanth
> <prashanth.dumpuri at vanderbilt.edu> wrote:
>   
>> Bill,
>>  Thanks for getting back to me. I have attached a pdf of the article and
>> have also listed below details about the article (just in case the
>> attachment does not go through).
>>
>> Multiple material marching cubes algorithm, Ziji Wu , John M. Sullivan Jr,
>> Volume 58, Issue 2  , Pages 189 - 207, International Journal for Numerical
>> Methods in Engineering.
>>
>> As for vtkImageMarchingCubes, I have not had success with it when trying to
>> create more than one marching cubes surface. For eg., I have an image volume
>> with a tumor in a human brain and I was trying to extract the tumor surface
>> and the brain surface. The average intensity of the tumor tissue is around
>> 100 (ranging from 1-200) and healthy brain tissue is anything with a
>> non-zero intensity value (ranging from 1-500). So when I ask vtk's
>> ImageMarchingCubes to extract tumor tissue with an isosurface value of 100,
>> I still get the brain tissue and not tumor. I know I'm missing something
>> here. Can you please point me in the right direction.
>> Thanks
>> Prashanth
>>
>>
>>
>> Bill Lorensen wrote:
>>     
>>> vtkDiscreteMarchingCubes generates multiple surfaces from labelled
>>> (segment) data. vtkMarchingCubes has always generated multiple
>>> isosurfaces from continuous data.
>>>
>>> Can you post a pdf of the journal article (assuming it is not a closed
>>> journal)? Or the name of the article. Hopefully the authors retained
>>> their rights to their own intellectual property and havede posted a
>>> copy (assuming they did not give up their rights to do so).
>>>
>>> Bill
>>>
>>>
>>> On Tue, Apr 8, 2008 at 3:09 PM, Prashanth
>>> <prashanth.dumpuri at vanderbilt.edu> wrote:
>>>
>>>
>>>       
>>>> All,
>>>> Does ITK have an implementation of the multiple material marching cubes
>>>> algorithm (International Journal of Numerical Methods in Engineering
>>>>         
>> 2003;
>>     
>>>> 58:189–207) ? By multiple material marching cubes, I mean creating
>>>> contiguous 3D surfaces for as many materials(iso-surfaces) requested
>>>>         
>> within
>>     
>>>> a single sweep of an image volume.
>>>> Thanks
>>>> Prashanth
>>>> _______________________________________________
>>>> Insight-users mailing list
>>>> Insight-users at itk.org
>>>> http://www.itk.org/mailman/listinfo/insight-users
>>>>
>>>>
>>>>
>>>>         
>>