Paraview Support for Computational Anatomy

• Implemented in conjunction with proposal 3D Shape Analysis for Computational Anatomy

• 

  Slicer scene with JHU Atlas.

• 

  Scene instantiated under Paraview.

• 

  ITK-based smoothing of surface mesh.

Key Investigators

• Kitware: Michel Audette, Luis Ibanez, Julien Finet, Andinet Enquobahrie, Stephen Aylward, Will Schroeder
• Johns Hopkins: Michael Bowers, Anthony Kolasny, Michael Miller
• CoSMo: Alex. Gouaillard

Objective

• Enable Paraview to support VTK-based visualization and ITK-based processing for 3D Shape Analysis Computational Anatomy.

Approach, Plan

• Anatomical components to be manipulated as a scene composed of MRML nodes, e.g.: VolumeNode, ModelNode, etc.
• Interaction possible between Slicer and Paraview.

Progress

A new library, KWScene, is being developed based on the MRML standard to support scene graph-type manipulation of volume and surface primitives for VTK-based visualization and ITK-based processing. This library will facilitate Paraview use in the computation pipeline and visualization of Computational Anatomy research at Johns Hopkins Center for Imaging Science. Reading and serialization is compatible with Slicer scenes.

Delivery Mechanism

Summer Project Week will be used to demonstrate to JHU CIS the work done so far at Kitware on KWScene, in relation to the data types and file formats routinely used in their Computational Anatomy research, and to develop a VTK-plugin mechanism for ITK-based processing on Paraview.

Update (June 24, 2010): we managed to imbed the JHU CIS group's ITK-based Laplace-Beltrami filter (http://www.cis.jhu.edu/software/laplace-beltrami/) into a VTK-plugin that runs on Paraview. We uncovered, along the way, a bug in the itk::QuadEdgeMesh, which was a useful exercise in itself, and we thank Alexandre Gouillard for his support in this area. Thanks to Julien Finet and Pat Marion of Kitware, who wrote a Python-based script that produces an animation from the surface harmonic outputs produced by the Laplace-Beltrami filter. Embedded here is a digital film of this animation produced with RecordMyDesktop.

Thanks to Curtis Lisle for converting the film to Quicktime mov format.

Click here for a quicktime video

References

• http://www.paraview.org/
• Mirza Faisal Beg, Michael I. Miller, Alain Trouvé, Laurent Younes: Computing Large Deformation Metric Mappings via Geodesic Flows of Diffeomorphisms. International Journal of Computer Vision 61(2): 139-157 (2005)
• Grenander, U. and Miller, M. I. 1998. Computational anatomy: an emerging discipline. Q. Appl. Math. LVI, 4 (Dec. 1998), 617-694.