ABSTRACT
This chapter describes the use of the Monte Carlo method to simulate nuclear medicine imaging systems, mainly the scintillation camera – SPECT (Single-Photon Emission Computed Tomography) and PET (Positron Emission Tomography) – systems that are major tools in nuclear medicine to produce images of activity distributions. The first part describes the principles behind the Monte Carlo method, in particular, how to select a stochastic variable from known probability distribution functions using uniform random numbers and, in more detail, how to sample photon interaction processes such as sampling photon path-length, photon interactions and scattering resulting in change in energy and direction. The improvement in calculation efficiency by implementation of various variance-reduction methods is also described. The second part describes in more detail two Monte Carlo codes, SIMIND and GATE, that for many years have been widely used for simulation of medical imaging. The potentials of these programs and how a user runs these programs are described by several explicit examples. Also described are applications where these codes have been useful, such as in image reconstruction, modelling of scatter and collimator septum penetration effects, and evaluation of pre-clinical imaging systems. Some perspectives, such as artificial intelligence approaches, SiPM-based SPECT/PET systems, or the electronically collimated Compton camera are also discussed.
