Dear Prof. Saba,

I have an idea how to possibly increase the number of projections calculated by your MCNP simulation. The eye lens is a small organ, so maybe a small CT section of the head is sufficient for your purpose (lens plus-minus 2-3 cm in cranio-caudal direction). In addition, in the photon energy range of a CT (X-rays up to 120 - 150 keV), the photo effect is the dominant interaction for generating dose in tissue (resulting electron path lenght < 1 mm in water) while scattering plays only a minor role. Using a small CT section would be sufficient from that point of view. Therefore you can maybe decrease the size of the projections which could increase your simulation power.

You can also try an iterative approach for the CT reconstruction using your about 80 projections (angular range of about 210 - 240° sufficient for a complete reconstruction), but as you said the image quality could be poor.

There is also some literature on CT reconstruction with limited number of projections, e.g.:

https://www.ncbi.nlm.nih.gov/pubmed/19692978

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2834797

I hope this helps you. Best regards

Dear Saba

As an alternative, we can use the interpolation to the sinogram data before reconstruction using FBP. Also, we can use the iterative reconstruction technique, e.g. SIRT, ART etc.