Tuberculosis (TB) is a very contagious infectious airborne sickness that has caused more deaths than any other disease. Despite being discovered over a century ago, it remains a global health threat, with high mortality rates due to delayed or missed diagnosis. TB is preventable and curable, but early detection is crucial to prevent transmission and improve treatment outcomes. Traditional methods of diagnosis, such as bacteriological tests and chest X-rays, can be time-consuming and prone to errors. Therefore, there is a need to develop accurate and efficient tools to detect TB at an early stage. Deep learning is a cutting-edge technique in the field of artificial intelligence that has the potential to be incorporated into computerized diagnosis systems for the purpose of performing automated TB detection in chest X-rays. The ultimate goal is to reduce patient waiting times and improve TB control. This article explores the challenges and opportunities involved in using deep learning for computer-generated diagnoses of active pulmonary tuberculosis in chest X-rays.