[Rtk-users] Fwd: some bugs in SART when running Reconstruction of Elekta Data

[Cyril Mory]

Hi Guangming,

I have just tested your command line and obtained some result, although 
it is indeed disappointing. SART generates strong artifacts on the 
borders of the image because some rays intersect only a few voxels on 
the borders and make the convergence very unstable. These artifacts are 
intense, so they mess up with the automatic choice of the window and 
level of the viewer.

My advice :
- Use "VV, the 4D slicer" to open your data (you should not see much at 
the beginning), and in the top right viewing pane, click in the center 
and hit the 'w' key. It will look at the pixels around the one you 
clicked, and adjust the window / level accordingly. Then you should see 
the reconstruction and the artifacts at the borders. Any other viewer 
with this "local windowing" option should do fine.
- If you want to reconstruct a full volume, not just one slice, you can 
use SART : the artefacts on the borders should be much less intense once 
your whole object is in the reconstructed volume
- If you want to reconstruct a single slice, or a few slices, use 
"rtkconjugategradient" instead. It does essentially the same as SART, 
but has a way steadier (although a bit slower) convergence. You can run 
it with the same options as SART, so all you have to do is replace 
"rtksart" by "rtkconjugategradient" in your command line examples (in 
particular, it runs fine on CUDA)

Do not hesitate to ask if you happen to have more questions.
Regards,

Cyril

On 11/20/2014 01:08 PM, Guangming Zang wrote:
> BTW, the Varian reconstruction has the same situation.I tested just 
> now that fdk work well while sart not.
>
> Best
> Guangming
> */
> /*
>
>
> ---------- Forwarded message ----------
> From: *Guangming Zang* <guangming.zang at kaust.edu.sa 
> <mailto:guangming.zang at kaust.edu.sa>>
> Date: 2014-11-20 14:21 GMT+03:00
> Subject: some bugs in SART when running Reconstruction of Elekta Data
> To: "rtk-users at public.kitware.com 
> <mailto:rtk-users at public.kitware.com>" <rtk-users at public.kitware.com 
> <mailto:rtk-users at public.kitware.com>>
>
>
> Hi,
> i ran the Reconstruction of Elekta Data example in wiki
>  (http://wiki.openrtk.org/index.php/RTK/Examples/ElektaReconstruction)
> when i use fdk to reconstruct the data, it work very well(can see the 
> slice). When ran the SART algorithm, however,i got the result with 
> same size of fdk algorithm's , but nothing can been seen in the slice.
> Here is the command i used:
> fdk:
>  rtkfdk --geometry elektaGeometry  --path 
> img_1.3.46.423632.135428.1351013645.166/ --regexp .*.his  --output 
> slice29.5.mha  --spacing 0.25,0.25,0.25 --dimension 1024,1,1024 
> --origin -127.875,29.5,-127.875
>
> SART(with different -f -b):
>  rtksart --geometry elektaGeometry  --path 
> img_1.3.46.423632.135428.1351013645.166/ --regexp .*.his  --output 
> slice29.6.mha  --spacing 0.25,0.25,0.25 --dimension 1024,1,1024 
> --origin -127.875,29.5,-127.875
>
> rtksart  -g elektaGeometry -f CudaRayCast  -b CudaVoxelBased   -t 1 -p 
> img_1.3.46.423632.135428.1351013645.166/ -r  .*.his -o slice_cuda.mha 
> --spacing 0.25,0.25,0.25  --dimension 1024,1,1024 --origin 
> -127.875,29.5,-127.875
>
> rtksart  -g elektaGeometry -f Joseph -b NormalizedJoseph   -t 1 -p 
> img_1.3.46.423632.135428.1351013645.166/ -r  .*.his -o slice.mha 
> --spacing 0.25,0.25,0.25  --dimension 1024,1,1024 --origin 
> -127.875,29.5,-127.875
> the results are the same, can not see anything.
>
> Can you help me to figure it out or give me some suggestions?
> Thanks for your time.
> */
> /*
>
>
>
> ------------------------------------------------------------------------
> This message and its contents, including attachments are intended 
> solely for the original recipient. If you are not the intended 
> recipient or have received this message in error, please notify me 
> immediately and delete this message from your computer system. Any 
> unauthorized use or distribution is prohibited. Please consider the 
> environment before printing this email.
>
>
> _______________________________________________
> Rtk-users mailing list
> Rtk-users at public.kitware.com
> http://public.kitware.com/mailman/listinfo/rtk-users

-- 
--
Cyril Mory, Post-doc
CREATIS
Leon Berard cancer treatment center
28 rue Laënnec
69373 Lyon cedex 08 FRANCE

Mobile: +33 6 69 46 73 79

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/rtk-users/attachments/20141120/d5bed16d/attachment-0009.html>