Detection and Quantification of TB Bacilli in Sputum Smear Images

Early and precise detection is essential to stop the spread of tuberculosis (TB) and enhance patient outcomes. This research aims to create an automated system that employs cutting-edge image processing and machine-learning approaches to identify tuberculosis (TB) bacilli in sputum smear microscopy pictures. An automated solution is required to increase diagnostic accuracy and efficiency because traditional diagnostic methods are labor-intensive, time-consuming, and prone to human mistakes. In the end, we optimize the best-performing model to reduce processing time while preserving diagnostic accuracy. Our goal is to assess the performance of several models in terms of training time, testing time, and accuracy. By enabling the use of sophisticated diagnostic technologies in environments with limited resources, the suggested method supports international efforts to combat tuberculosis.