Tuberculosis (TB) is one of the worst lung infections caused by bacteria
called Mycobacterium tuberculosis. It is a disease that can be controlled in cases where
it is quickly identified and treated. As sputum microscopy is a simple and low-cost
approach, most countries use this technique as an initial step in the diagnosis of TB.
For this, the patients' morning sputum is collected and submitted to the Ziehl Neelsen
staining procedure before the examination. Then a digital microscope is used, where
the images of the sputum slides can be recorded for analysis. In this context, numerous
research projects have been developed in the field of computer-aided detection of
mycobacteria for pulmonary tuberculosis. The survey of these works was discussed
here in this work, as well as their limitations.