Tuberculosis is an important infectious disease caused by a bacterium and spread all over the world. Although some individuals develop a severe active tuberculosis infection (ATB), most of the individuals evolve into the asymptomatic Latent Tuberculosis Infection (LTBI), through the formation and control of granuloma inside a pulmonary secondary lobule. In such case, the growth of the lesion is restricted inside one of the secondary lobules. There are around \(10^3\) secondary lobules in a human lung. The development of the lesion inside the lung can be modeled by a set of reaction-diffusion equations involving different types of cells. However, here a reduced model has been developed to study the evolution of an initial infection until it is encapsulated, both in 2D and in 3D. For the 2D case, a sensitivity analysis has been performed to study the influence on the model’s parameters to the final radius of the granuloma and the encapsulation time. For the 3D case, unstable and stable approaches have been considered and their encapsulation processes compared. The model was first implemented in a cube and afterwards extended to an irregular shaped secondary lobule. In the near future, we plan to apply such model to a virtual lung to first reproduce the results of a lung-scale agent-based model that fits the experimental results of the tuberculosis spreading along the whole lungs, and next to be able to understand the process involved in the differences between the active and the latent versions of the tuberculosis infections.