CT image voxels simulation

This is a follow up questions to what was discussed earlier in this link Treatment planning for radiation therapy - #3 by horvathd
I’ve got only CT image with no Rtplan and been able to import and convert HU to voxel and finally voxel.vxl is generated. To run usual FLUKA source, I’ve added a new voxel card then from dicom tab I’ve added USRBIN. Everything looks great even geometry has been checked with no errors. When I run the simulation it finishes soon and error file shows no stars have been generated. I’v tried to implement the SOPB subroutines to simulate proton beam but it also produces the same output.
the input file is attached.
Any help would be appreciated

Dear @abdalmalki,

Could you please provide the following files:

  • Flair and input
  • Voxel file


Here you go
photon.flair (2.0 KB)
photon.inp (1.4 KB)
voxel file is huge so that I uploaded it on dropbox, please find it at

if the above link doesn’t work, please try the following one


Dear @abdalmalki,

The problem is in the particles source. In the way it is, it is not hitting your geometry :slight_smile:
Here you can see the particles trajectories after using the USERDUMP card.

That’s it. If you put properly the BEAMPOS card it should be ok.

Just two comments in order to avoid this kind of problems.

  1. As I said before, you can use the USERDUMP card in the following way:

Screenshot from 2021-03-29 21-16-15

There you will get the tracks of all the particles generated in the simulation (be careful, the outputs generated here are quite big)

  1. And a simple check, in the tab ‘Geometry’ in Flair (where you see the geometry), there you can observe the direction of the particles source (a violet arrow) as it is shown in the first screenshot that I added here.


1 Like

Dear Abdul,

in addition to André’s answer, there are some easier ways to visualize the beam path in FLUKA:

  1. You can use a smple USRBIN scoring with BEAMPART particle, or
  2. Click the geoviewer’s BEAM object.

For more information see slide 15-16 of this lecture:


A post was split to a new topic: Using CORRFACT with CT voxels