Key points are not available for this paper at this time.
Objectives Worldwide, tuberculosis (TB) continues to be an important cause of human morbidity and mortality, particularly in developing countries where drug surveillance and rapid detection of resistance to anti-TB drugs is uncommon and the lack of proper healthcare systems often leads to incomplete treatment and spread of MTB strains. The present paper discusses the potential role of artificial intelligence (AI) in improving TB diagnosis, management, and control in African countries, where conventional diagnostic approaches remain predominant and often insufficient. Methods A narrative review of current challenges in TB management in African settings was conducted, focusing on limitations of existing diagnostic tools and the emerging contributions of AI-based technologies in healthcare. The review demonstrates how machine learning algorithms and computer-aided systems could be integrated into TB programs to enhance clinical decision-making and surveillance. Results Conventional TB diagnostic methods such as the tuberculin test, radiography, and microscopic examination show limited accuracy and efficiency, contributing to ongoing transmission and poor treatment outcomes in many African countries. AI innovations have demonstrated improved performance in disease detection and prediction across various health domains, offering time-saving, resource-efficient, and scalable solutions. Applied to TB, AI could support clinicians in diagnosis, forecast treatment outcomes, and strengthen public health strategies aimed at controlling MTB spread. Conclusion AI holds significant promise for enhancing TB control efforts in African countries by improving diagnostic precision, clinical decision-making, and surveillance capacities. Integrating AI into national TB programs could promote more effective and efficient disease management, ultimately contributing to reduced morbidity and mortality.
Chaoui et al. (Wed,) studied this question.