[Vv] clitkDicomRTStruct2Image and VTK5/VTK6

[Thomas BAUDIER]

Hi Neree.

Thank you to report this bug. I reproduced it on my computer and I 
started to look for the problem last Friday. For the moment I don't know 
if the problem come from vv or vtk.
I just have a question: Do you know which version is correct between 
vtk5 and vtk6?

For your second point, the difference is due to the RTStruct management:
  - In 3DSlicer, the RTStruct is converted into a Mesh
  - In vv, the RTStruct is converted into an Mesh and next an image. So 
during the conversion between Mesh and image, vtk has to interpolate on 
pixels to create the binary ROI.

Thank you
Thomas

On 2017-05-12 15:56, Neree PAYAN wrote:
> Hello,
> 
> I’m using VV software for Positron Emission Tomography image
> processing. I have DICOM PET images with the associated ROIs (=
> RT-Struct DICOM format). I wanted to use the clitkDicomRTStruct2Image
> command to export the ROIs in binary image (nifty format), and I met
> two problems:
> 
> 1. I used 2 different versions of vtk (vtk5 and vtk6) for this work,
> and the result was not the same according to the versions. There is a
> shift of one slice between the two output binary images (for the exact
> same input image (PET + RT-Struct)).
> I visualized and compared the 3 ROIs images (the RT-Struct and the two
> nifty images) on the 3DSlicer software, and the offset remains: the
> binary image made with the vtk5 version matches with the RT-Struct
> DICOM image, but the binary image made with vtk6 is shifted compare to
> the others.
> 
> 2. Moreover, comparing the ROIs images on 3DSlicer, I also noticed
> that the shape was not the same between the RT-Struct and the binary
> images converted with the clitk tool.
> 
> Have you got any idea why I have such differences ?
> You can find the PET Image with the RT-Struct associated and the
> resulting binary ROI images on this download link:
> https://cloud.u-bourgogne.fr/index.php/s/IV9t9ksgMTHjhxD .
> 
> Thank you in advance for your help,
> Regards,
> 
> Nérée PAYAN
> PhD Student, University of Burgundy
> 
> LE2i Laboratory, Health and Medical Imaging Department,
> UFR Sciences et Techniques,
> Allée Alain Savary,
> 21000 Dijon, France
> 
> nerea.payan at u-bourgogne.fr
> Tel : +33 (0)6 68 82 54 67
> _______________________________________________
> vv mailing list
> vv at creatis.insa-lyon.fr
> http://www.creatis.insa-lyon.fr/mailman/listinfo/vv