Speaker
Description
The analysis of positronium (Ps) annihilation lifetime spectra requires multi-component modelling, constituted by exponentially modified Gaussian functions. A direct curve fitting is an ill-posed problem and needs either deconvolution or multi-stage constrained minimisation. In Ps lifetime imaging (PLI), a technique that targets local Ps lifetime distribution, an additional challenge emerges – the low number of counts per spectrum.
In this talk, we review the application of non-linear Bayesian fitting in PLI, based on the prior knowledge about the parameters of the Ps lifetime model. The available software implementations shall be compared, with the preliminary results for the simulated data.