Particulate air pollution is associated with respiratory and cardiovascular morbidity and mortality. Recent studies investigated whether and to which extent inhaled ultrafine particles are able to translocate into the bloodstream in humans. However, their conclusions were conflicting. We developed a physiologically based kinetic model for (99m)technetium-labelled carbon nanoparticles (Technegas). The model was designed to analyse imaging data. It includes different translocation rates and kinetics for free technetium, and small and large technetium-labelled particles. It was calibrated with data from an experiment designed to assess the fate of nanoparticles in humans after inhalation of Technegas. The data provided time courses of radioactivity in the liver, stomach, urine, and blood. Parameter estimation was performed in a Bayesian context with Markov chain Monte Carlo (MCMC) techniques. Our analysis points to a likely translocation of particle-bound technetium from lung to blood, at a rate about twofold lower than the transfer rate of free technetium. Notably, restricting the model so that only free technetium would have been able to reach blood circulation resulted in much poorer fits to the experimental data. The percentage of small particles able to translocate was estimated at 12.7% of total particles. The percentage of unbound technetium was estimated at 6.7% of total technetium. To our knowledge, our model is the first PBPK model able to use imaging data to describe the absorption and distribution of nanoparticles. We believe that our modeling approach using Bayesian and MCMC techniques provides a reasonable description on which to base further model refinement.