Evolution of 3D X-Ray Dose Computation Algorithms

Abstract

Radiation treatment planning of individual cancer patients relies on the accurate computation of dose distributions in irradiated tissue. Inaccurate dose maps have the potential to mislead clinical decision-making and compromise the balance between effective tumour control and side effects in surrounding normal tissue. In the context of this conference, 3D dosimetry is important for the experimental validation of computed dose distributions. Dose computation methods for external beams of high energy x rays have evolved over the past decade with computer simulation models more closely aligned with the fundamental physics of x-ray scattering and absorption in heterogeneous tissue. In this article, we first present a historical review from a Canadian perspective, followed by an introductory intuitive description of contemporary algorithms used in clinical treatment planning: (1) Convolution-superposition algorithm fundamentally based on the Green’s function method; (2) Stochastic Monte Carlo simulation of x-ray interactions in tissue, and (3) Deterministic numerical solution of a system of Boltzmann transport equations. In principle, all these methods solve the same problem of predicting x-ray scattering and absorption in heterogeneous tissue. However, the mathematical tools differ in their approach and approximations to achieve sufficient speed for routine clinical application. In the conclusion of this article, the evolution of 3D x-ray dose computation is summarized, in terms of accuracy and computational speed.