Dear Ilker,

this question was posed 4 years ago here in RG: https://www.researchgate.net/post/How_can_you_convert_CT_images_to_MCNP_transport_code_format_for_radiation_transport_simulations_in_radiotherapy_planning

People (Robert Zamenhof and Telma Cristina Ferreira Fonseca) suggested the following resources:

I do not know what exactly you want to do with the converted CT data: radiotherapy planning?

Depending on which information you need, I suggest you preprocess your images with an opensource library such as https://pydicom.github.io/pydicom/stable/ which offers a lot of possibilities in using metadata and simple image analysis algorithms and then export them as as MNCP (you'll have to read the documentation on the MNCP subformats).

Good luck!

Aldo

Thanks for Your response Aldo ! Sure. I was aware of that question on RG but as you said, it has been four years since it was last asked and I just wanted to see if there were more recent developments. We'll need to use these images both for planning and in the future (if our ideas succeed), also for range and dose verification in particle therapy. I'll have a look at these references anyway to see if these could help. Pydicom could definitely help in this case. In any case, it seems we'll have to come up with a partial in-house solution to this.

Just a follow-up on this. Thank you again Aldo, for recommending the pydicom library. I'd dare say it indeed did the job nicely for us. We now have a pydicom-based routine that can convert DICOM images to MCNP input files. We have used MCNP macrobodies to do the job such that the input files end up being large depending on the voxel sizes. I guess the size of input files could be further compressed using the lattice approach. We'll look into that. Here is the GitHub rep. with the mcnp-parser : https://github.com/ivmmoreira/dicom-mcnp-parser

This is still a work-in-progress as some testing/debugging will be needed to confirm that everything works as expected. Initial images look OK though.