Simulation of Binary Methods for the Microdosimetric Analysis of Cell Survival after Alpha-Particle Irradiation: Ability to Distinguish between Different Models

Abstract

Analysis of cell survival after alpha-particle irradiation must account for the distribution in the amounts of energy deposited in each cell nucleus. Microdosimetric computations are usually used to determine these distributions. Irradiation with microbeams and other modern techniques has made these computations unnecessary for certain cell geometries. These techniques allow the survival of individual cells to be correlated with the amount of radiation delivered to individual cell nuclei. However, to maintain the individuality of data generated for each cell, new methods of analysis are required. In this study, we propose the use of binary methods. Each cell is regarded as a Bernoulli trial with a different probability for success (colony formation). Parameter values of the survival model are chosen to maximize the likelihood of the observed outcome. To evaluate this method, simulated data for 500, 5000 and 50,000 cells irradiated by alpha particles are analyzed along with the associated outcome for four different cell survival models. Each survival model has a different dependence on the radius of the cell nucleus. These results indicate that the model that was simulated has the highest likelihood value in all cases. However, the ability to distinguish between competing models is present only for a larger numbers of cells.