Development of Monte Carlo models for the optimisation of positron emission particle tracking experiments

Positron emission particle tracking (PEPT) is a non-invasive technique used to measure the three-dimensional position of positron-emitting tracers. PEPT is useful for studying myriad industrial and/or scientific systems which often are optically inaccessible. However, when running an experiment, often little attention is paid to optimising the tracers, detectors, algorithms, and experimental procedures. As a result, trajectories can be degraded leading to inefficient use of resources. To address this opportunity, Monte Carlo simulations are employed to model experiments and predict the tracer activity, detector geometry, and algorithm parameters that will produce the best trajectory resolution possible and even determine whether an experiment is feasible or how long the experiment should be conducted to reduce uncertainty to an acceptable degree. Importantly, this simulated work can be conducted prior to experiments. In this thesis, a general procedure for simulating PEPT experiments is described which can be applied to any PEPT experiment. The results of this work demonstrate that not only is this method able to produce realistic synthetic PEPT data, but allows, for the first time, a quantitative comparison of PEPT algorithms, the ability to optimise experiments, and to develop new PEPT methodologies using information difficult or impossible to extract from real experiments.