Tuberculosis remains an urgent issue on the urban health agenda, especially in low-and middle-income countries. There is a need to develop and implement innovative and effective solutions in the tuberculosis diagnostic process. In this article, We describe the importance of artificial intelligence as a strategy to address tuberculosis control, particularly by providing timely diagnosis. Besides technological factors, the role of socio-technical, cultural and organizational factors is emphasized. The eRx tool involving deep learning algorithms and specifically the use of convolutional neural networks is presented as a case study. eRx is a promising artificial intelligence-based tool for the diagnosis of tuberculosis; which comprises a variety of innovative techniques involving remote X-ray analysis for suspected tuberculosis cases. In-novations based on artificial intelligence tools can optimize the diagnostic process for tuberculosis and other communicable diseases.