Deterministic and Monte Carlo radiation transport in Computed Tomography dose estimation

Joshua Hykes, North Carolina State University

The popularity of Computed Tomography (CT) imaging has increased dramatically in the past two decades. Given the relatively large radiation dose imparted during a CT scan, accurately estimating the dose is critical for safe and effective use of the technology, especially for pregnant patients. Computational models provide much greater flexibility and reduced costs for this purpose than experimental methods. I seek to develop a deterministic model of the gamma radiation transport during a CT scan. Employing a deterministic approach instead of the more typical Monte Carlo method has a number of advantages. The most obvious is the greater computational efficiency of the method, but it also provides greater spatially-detailed flux information without the statistical uncertainties of Monte Carlo. However, the deterministic method lacks some of the physics of Monte Carlo, notably the transport of secondary electrons and continuous energy. I present a number of comparisons of the results of TORT and MCNP5, production-quality deterministic and Monte Carlo packages, respectively, on a simple geometrical model to demonstrate that the two methods have reasonable agreement. With this confidence, a validation of the deterministic model against experimental data is possible.

Abstract Author(s): Joshua Hykes