[Insight-users] Problems with itkDeformableMesh3DFilter

[Sven Olaf Lüttmann]

Thanks for the quick reply Bill,
if I get your point right, your saying that because of the different coordinate system the y-axis is 
flipped, so for my imagination my model would be upside down. But that should not bring up any 
computational issues, should it?

I tried a simple RegularSphere in itk (RegularSphereMeshSource) as initial mesh for the segmentation 
process. Same problem.
I debugged the code as far as possible and it came up that the process crashes in the function 
"ComputeDt()" in the DeformableMesh3DFilter class line 331.

The Array "m_K" is not filled properly with the stiffness parameters which should be done in the 
function "SetMeshStiffness()" in the same class. This function skips the while loop (see below) in 
line 288 which probably causes the process to crash:

while (celldata != myCellData->End())
     {
     const double x = celldata.Value();
     m_K[j] = m_StiffnessMatrix+((int) x);
     ++celldata;
     j++;
     }

Hope that helps for further assistance.

Thanks very much

Cheers
  Sven



*Fachhochschule Dortmund*
*University of Applied Sciences and Arts*

Sven Olaf Lüttmann
B.Sc. Medizinische Informatik

FB Informatik, Medizinische Informatik
Emil-Figge-Str. 42
44227 Dortmund

Raum: EFS 42 B.1.21

Fax: 0231 / 755-6725
E-Mail: sven.luettmann at fh-dortmund.de <mailto:sven.luettmann at fh-dortmund.de>
Web: http://www.fh-dortmund.de

On 13.10.2011 13:59, Bill Lorensen wrote:
> ITK and VTK have different coordinate systems. The VTK y-axis is flipped. Perhaps this is causing 
> issues.
>
> On Thu, Oct 13, 2011 at 7:34 AM, Sven Olaf Lüttmann <sven.luettmann at fh-dortmund.de 
> <mailto:sven.luettmann at fh-dortmund.de>> wrote:
>
>     Dear ITK-Users,
>
>     I'm experiencing problems implementing a deformable segmentation approach. I'm using the
>     itkDeformableMesh3DFilter.
>
>
>     Here is what I do specifically:
>
>       1.
>
>           Let the user draw contours with vtkContourWidget
>
>       2.
>
>           Connect these contours to form a model as vtkPolyData
>
>       3.
>
>           Convert the vtkPolyData to itk::Mesh using the code mentioned in the Auxiliary folder in
>           the InsightApplications (vtkPolyDataToITKMesh.cxx)
>
>       4.
>
>           Use that mesh as initial mesh for the segmentation process
>
>
>     The problem is when I use my generated mesh the segmentation process crashes. When I use
>     exactly the same code as in the example, means using the BinaryMask3DMeshSource Filter to
>     create the mesh, everything works fine.
>
>
>     Are there any known issues converting vtkPolyData into itk::mesh.
>
>
>     Thanks in advance
>
>     Best regards
>
>         Sven Lüttmann
>
>
>     -- 
>     *Fachhochschule Dortmund*
>     *University of Applied Sciences and Arts*
>
>     Sven Olaf Lüttmann
>     B.Sc. Medizinische Informatik
>
>     FB Informatik, Medizinische Informatik
>     Emil-Figge-Str. 42
>     44227 Dortmund
>
>     Raum: EFS 42 B.1.21
>
>     Fax: 0231 / 755-6725
>     E-Mail: sven.luettmann at fh-dortmund.de <mailto:sven.luettmann at fh-dortmund.de>
>     Web: http://www.fh-dortmund.de
>
>     _____________________________________
>     Powered by www.kitware.com <http://www.kitware.com>
>
>     Visit other Kitware open-source projects at
>     http://www.kitware.com/opensource/opensource.html
>
>     Kitware offers ITK Training Courses, for more information visit:
>     http://www.kitware.com/products/protraining.html
>
>     Please keep messages on-topic and check the ITK FAQ at:
>     http://www.itk.org/Wiki/ITK_FAQ
>
>     Follow this link to subscribe/unsubscribe:
>     http://www.itk.org/mailman/listinfo/insight-users
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/pipermail/insight-users/attachments/20111013/871ab3c5/attachment.htm>