Dear Miguel,
indeed this is weird. Can you run the model through a cleaner and a
connectivity, this way:
vmtksurfacetriangle -ifile yoursurface.vtp --pipe vmtksurfaceconnectivity
-ofile yoursurface_c.vtp
and then run the algorithm on it? If all else fails, can you make the model
available to me for a
quick test?
BTW, cool work
Luca
On Nov 30, 2012, at 3:23 PM, Bernabeu Llinares, Miguel wrote:
> Dear VMTK users,
>
> I continued working on the project I described in my e-mail on 30 Sep 2012
> (see below). Following Arjan's advice I managed to get vmtkcenterlinemodeller
> to accurately resolve the vascular network.
>
> I'm currently processing a new image I obtained from my experimental
> collaborators and the same workflow I used in the past is failing at the
> vmtknetworkextraction step. I'm attaching two Paraview snapshots: one with
> the surface mesh from which I'm trying to obtain centrelines and a second one
> with the output generated by vmtknetworkextraction.
>
> I guess there's some singularity in the surface mesh that is making the
> vmtknetworkextraction algorithm fail (given that it has worked in the past
> for similar models). I'm unfortunately pretty clueless about what might be
> the culprit. Note that I've tried opening the surface mesh at the different
> points to no improvement.
>
> Any help would be greatly appreciated!
>
> Best wishes,
> Miguel
>
> <network.png>
> <centrelines.png>
>
> On 30 Sep 2012, at 13:03, Arjan Geers wrote:
>
>> Dear Miguel,
>>
>> It seems that the output image of vmtkcenterlinemodeller doesn't resolve all
>> the details of the vascular network. You can improve the resolution of this
>> image by changing the 'dimensions' parameter. Example with default values:
>>
>> vmtkcenterlinemodeller -ifile surface_decimated095_clipped_centerlines.vtp
>> -radiusarray Radius -dimensions 64 64 64 --pipe vmtkmarchingcubes -ofile
>> surface_decimated095_clipped_centerlines_tubed.vtp
>>
>> Since the centerlines lie in the xy-plane, you'll probably need less voxels
>> in the z-direction.
>>
>> Hope this helps,
>>
>> Arjan
>>
>> On Sun, Sep 30, 2012 at 11:18 AM, Bernabeu Llinares, Miguel
>> <miguel.berna...@ucl.ac.uk> wrote:
>> Dear VMTK users,
>>
>> I'm using VTMK 1.0.0 to generate a 3D surface mesh from a single 2D image of
>> a complex vascular network. A similar application was previously discussed
>> in the mailing list:
>>
>> http://sourceforge.net/mailarchive/forum.php?thread_name=6DEB2248-5927-48CC-AB2F-E24662DF4A05%40orobix.com&forum_name=vmtk-users
>>
>> I followed Luca's advice of replicating the image to generate a volume. From
>> then, I used a combination of marching cubes and vmtknetworkextraction to
>> obtain the centrelines of the model and its associated maximum inscribed
>> sphere radius. What I would like to do next is to use vmtkcenterlinemodeller
>> and marching cubes to fit tubes around the centrelines to obtain a network
>> of cylindrical section (as opposite to the network of square ducts coming
>> out of the first marching cubes step). Something like:
>>
>> vmtknetworkextraction -ifile surface_decimated095_clipped.vtp -ofile
>> surface_decimated095_clipped_centerlines.vtp
>>
>> vmtkcenterlinemodeller -ifile surface_decimated095_clipped_centerlines.vtp
>> -radiusarray Radius --pipe vmtkmarchingcubes -ofile
>> surface_decimated095_clipped_centerlines_tubed.vtp
>>
>> Unfortunately the result is not what I expected: many vessels collapsed
>> together and the rich topology is mostly lost. Please see the two images
>> attached with the results of the two previous commands.
>>
>> Any input is very much appreciated!
>>
>> Best wishes,
>> Miguel
>>
>>
>> <centerlines_with_radius.png>
>>
>> <tubes_around_centerlines.png>
>> --
>> Dr Miguel O. Bernabeu
>> 2020 Science Research Fellow (http://www.2020science.net)
>> Centre for Computational Science, University College London
>> CoMPLEX, University College London
>>
>>
>> ------------------------------------------------------------------------------
>> Everyone hates slow websites. So do we.
>> Make your web apps faster with AppDynamics
>> Download AppDynamics Lite for free today:
>> http://ad.doubleclick.net/clk;258768047;13503038;j?
>> http://info.appdynamics.com/FreeJavaPerformanceDownload.html
>> _______________________________________________
>> vmtk-users mailing list
>> vmtk-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/vmtk-users
>>
>>
>
> --
> Dr Miguel O. Bernabeu
> 2020 Science Research Fellow (http://www.2020science.net)
> Centre for Computational Science, University College London
> CoMPLEX, University College London
>
> ------------------------------------------------------------------------------
> Keep yourself connected to Go Parallel:
> TUNE You got it built. Now make it sing. Tune shows you how.
> http://goparallel.sourceforge.net_______________________________________________
> vmtk-users mailing list
> vmtk-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/vmtk-users
------------------------------------------------------------------------------
Keep yourself connected to Go Parallel:
BUILD Helping you discover the best ways to construct your parallel projects.
http://goparallel.sourceforge.net
_______________________________________________
vmtk-users mailing list
vmtk-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/vmtk-users
