Deep Learning Techniques for Tuberculosis Detection on Chest X-Rays in Low-Resource Settings: A Survey of Opportunities and Challenges

Tuberculosis (TB) remains a significant public health challenge, particularly in low-resource settings, where limited access to diagnostic tools and healthcare infrastructure exacerbates the burden of the disease. This survey critically examined the application of deep learning techniques for TB detection using chest X-rays, focusing on the opportunities and challenges specific to low-resource environments. This study reviews state-of-the-art deep learning models, such as Convolutional Neural Networks (CNNs), transfer learning frameworks, and hybrid approaches, to assess their effectiveness in enhancing diagnostic accuracy, scalability, and cost-efficiency. The key challenges in implementing these models include the scarcity of annotated data, limited computational resources, and the need for model adaptability across diverse populations. These findings underscore the transformative potential of deep learning in improving TB detection, particularly through automation, integration with telemedicine, and deployment of lightweight, efficient models suitable for resource-constrained settings. This survey highlights the critical need for continued research and innovation to overcome existing barriers and fully realize the benefits of deep learning in TB diagnostics.