A 3D computational model for understanding tuberculosis lesions dynamics in lungs

Abstract

Tuberculosis (TB) is an infectious bacterial disease caused by Mycobacterium tuberculosis (Mtb), which most commonly affects the lungs. In healthy people, an infection with Mtb often causes no symptoms, remaining controlled as a non-contagious latent tuberculosis infection. World Health Organization estimates that one third of the world population is already infected by this bacillus. From those, a 10% will probably develop an active disease the next decade. Nowadays, over 1 million people die annually because of an active TB. The mechanisms that maintain a latent infection for a few years or that make it evolving towards an active disease are not fully understood, yet. In a previous work, the dynamics of TB lesions during an active disease in mice was described by an Agent-Based Model (ABM). This model accounted for the growth, coalescence and proliferation of lesions, showing that the most important mechanism for lesions growth during the active disease was coalescence. In a later work, the dynamics of lesions during a latent infection in minipigs was tackled by implementing a revised version of the previous ABM into a computational model of the bronchial tree. The model was fed with Computed Tomography scan data from latent infection in minipigs. In this case, the model showed that the proliferation of lesions through the bronchial tree was essential for maintaining the latent infection. In this Master thesis we propose a first approach on the evolution of a latent tuberculosis infection into an active disease. The parameter space will be explored trying to elucidate which is the role of each mechanism on the trigger for the disease