Our research is motivated by the rapidly-evolving outbreaks of rare and fatal infectious diseases, for example, the severe acute respiratory syndrome (SARS) and the Middle East respiratory syndrome. In many of these outbreaks, main transmission routes were healthcare facility-associated and through person-to-person contact. While a majority of existing work on modelling of the spread of infectious diseases focuses on transmission processes at a community level, we propose a new methodology to model the outbreaks of healthcare-associated infections (HAIs), which must be considered at an individual level. Our work also contributes to a novel aspect of integrating real-time positioning technologies into the tracking and modelling framework for effective HAI outbreak control and prompt responses. Our proposed solution methodology is developed based on three key components – time-varying contact network construction, individual-level transmission tracking and HAI parameter estimation – and aims to identify the hidden health state of each patient and worker within the healthcare facility. We conduct experiments with a four-month human tracking data set collected in a hospital, which bore a big nosocomial outbreak of the 2003 SARS in Hong Kong. The evaluation results demonstrate that our framework outperforms existing epidemic models for characterizing macro-level phenomena such as the number of infected people and epidemic threshold.