Dear Jean, from the picture you attached it is not entirely clear to me how much contrast is there within the aorta. The window/level is too wide: bones, which should be clearly visible, do not stick out either. In any case, I've had some luck with poorly enhanced CT's with a very noisy lumen in the past. In that case I downsampled the image using vmtkimagereslice (e.g. reduce in-plane resolution by a half, using cubic interpolation), or you can also try smoothing the image first (using vmtkimagesmoothing with a dimensionality 2 and a sigma equal to your downsampled resolution) and then downsampling.
As for the cannula, if you generate the mesh for the cannula separately you can append the two surfaces together using vmtksurfaceappend (which doesn't connect them topologically, it just collects the two surfaces in a single dataset). The only gotcha I see towards meshing is that you have to close your domain at the inlets/outlets embedding the cannula in them, which is something that vmtk would not allow you to do out of the box. An alternative is to have the level set image for the aorta (say, the aorta_ls.vti file), the level set of the cannula, and subtract them. This way you can cut out the cannula volume from the aorta volume and have a closed surface in output that you can mesh, even using vmtkmeshgenerator. Hope this helps. Luca On Feb 14, 2011, at 8:21 PM, Jean Bonnemain wrote: > Dear all, > > I'm using vmtk to generate a mesh of the thoracic aorta. Since the DICOM > images I use are not ideal, i.e. this is not in the full contrast enhanced > phase, it is difficult to generate a surface that corresponds to the aorta: > either the surface is "dented", because the threshold is too high, or the > surface extends to the other adjacent vessels. Is there a possibility to > avoid these problems? I join to this e-mail a picture of a slice. > > Another question : I would like to create a mesh of a cannula inserted into > the aorta. Is there a way in vmtk, starting from a normal aorta, to modify > the surface and add "the cannula" (mainly a tube)? > > Best regards, > Jean Bonnemain > <Thoracic_Aorta.png>------------------------------------------------------------------------------ > The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE: > Pinpoint memory and threading errors before they happen. > Find and fix more than 250 security defects in the development cycle. > Locate bottlenecks in serial and parallel code that limit performance. > http://p.sf.net/sfu/intel-dev2devfeb_______________________________________________ > vmtk-users mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/vmtk-users ------------------------------------------------------------------------------ The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE: Pinpoint memory and threading errors before they happen. Find and fix more than 250 security defects in the development cycle. Locate bottlenecks in serial and parallel code that limit performance. http://p.sf.net/sfu/intel-dev2devfeb _______________________________________________ vmtk-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/vmtk-users
