Enhancing Tuberculosis Detection: A Review of Optimization Algorithms in Medical Imaging

Abstract

Tuberculosis (TB) remains a significant global health challenge, particularly in South Africa, where it leads to substantial mortality each year. The disease is caused by the bacterium Mycobacterium tuberculosis, primarily affecting the lungs but potentially impacting other areas of the body as well. This study reviews various optimization algorithms that enhance machine learning and deep learning models for the detection of TB in medical imaging. Our aim is to provide valuable insights for researchers, practitioners, and decision-makers by highlighting the application of these optimization techniques throughout the stages of TB diagnosis. Additionally, we identify emerging areas where optimization algorithms could further improve TB detection, ultimately aiding in the fight against this disease and reducing mortality rates. We specifically showcase the effectiveness of seven optimization algorithms in clustering-based image segmentation, with results indicating that the top three performers are the Surrogate Algorithm, Particle Swarm Optimization with Pattern Search (PSOPS), and Genetic Algorithm with Pattern Search (GAPS).