[Gate-users] Question about GPU tracking on CT simulation

[邢潇丹]

Dear Gate users，
  Recently I’ve been trying to simulate CT based on gate. To accelerate my simulation, I attached GPU tracking inside my codes. When I ran the example provided by gate, in which the source is mono-energetic, everything went well and I could get a fine CT image. But then I tried to change the source energy into an arbitrary spectrum, the beam seemed to be absorbed completely by the phantom. The CT image was all black without any particles. As long as I disabled the GPU tracking actor, the CT image output turned fine again. I googled it and find this——
  http://hypernews.slac.stanford.edu/HyperNews/geant4/get/Multithreading/49.html
  It seems that I’m not the only one being trapped by this problem. And that really puzzles me.
  Thanks a lot for your help. Any reply will be truly appreciated. 

--

Xiaodan Xing
The Department of Physics
Fudan University