The comet assay was recently applied for the first time to test the genotoxicity of micrometric stainless steel and cement particles, representative of those produced in the dismantling of nuclear power plants. A large dataset was obtained from in vitro exposure of BEAS-2B lung cells to different concentrations of hydrogenated (non-radiative control) and tritiated particles, to assess the impact of accidental inhalation. Starting from the distributions of the number of nuclei scored at different extent of DNA damage (% tail DNA values), we propose a new comet data treatment designed to consider the inhomogeneity of the action of such particles. Indeed, due to particle behavior in biological media and concentration, a large fraction of cells remains undamaged, and standard averaging of genotoxicity indicators leads to a misinterpretation of experimental results. The analysis we propose reaches the following goals: genotoxicity in human lung cells is assessed for stainless steel and cement microparticles; the role of radiative damage due to tritium is disentangled from particulate stress; the fraction of damaged cells and their average level of DNA damage are assessed separately, which is essential for carcinogenesis implications and sets the basis for a better-informed risk management for human exposure to radioactive particles.