The Tuberculosis, a bacterial infectious disease, is mostly spread by aerosol transmission mainly affects the lungs. Being the second greatest infectious cause of mortality, this disease remains a major global challenge. As per the WHO survey, this disease has infected about 25% of the world’s population. Better understanding of the infection, appropriate diagnosis and timely treatment are required to combat tuberculosis over the world. Though the amount of bacteria in the sputum and cough frequency forms the two major indicators of Tuberculosis infection, very few studies have been made in identifying the patterns and frequency of cough. This review paper summarizes the different research works carried out to identify and analyze tuberculosis using cough as one of the primary indicators. This review analyzes the various cough datasets, cough counting tools and analysis, validation of cough screening machine-learning algorithms, artificial intelligence models to emphasize on the importance of cough frequency in tuberculosis infection detection.