Advances in Tuberculosis Diagnosis: Molecular, AI-Enabled, and Decentralized Approaches

Tuberculosis (TB) remains one of the leading causes of mortality from a single infectious agent worldwide and continues to pose a major public health challenge. This burden is further exacerbated by the emergence and spread of drug-resistant strains, which complicate treatment and increase transmission risks. Early and accurate diagnosis is essential for timely initiation of therapy, effective disease control, and prevention of resistance development. However, conventional diagnostic approaches, including clinical assessment, microbiological testing, radiological imaging, and immunological methods, still face important limitations such as delayed results, variable sensitivity, and dependence on laboratory infrastructure.Recent advances in digital health and molecular diagnostics are reshaping the TB diagnostic landscape. Artificial intelligence (AI)-driven tools, particularly those based on imaging and pattern recognition, have improved the speed, accuracy, and accessibility of TB detection, especially in resource-limited settings. These approaches may also help identify TB in individuals who are unable to produce adequate sputum samples. At the same time, innovations in nucleic acid amplification testing, line probe assays, targeted next-generation sequencing, CRISPR-based methods, and near-patient platforms are expanding the possibilities for rapid diagnosis and more precise detection of drug resistance.This Research Topic aims to evaluate and bring together these emerging tools and strategies within a clinically and programmatically relevant framework for TB diagnosis. It will highlight how molecular methods, AI-enabled tools, sequencing approaches, and decentralised point-of-care strategies can improve diagnostic accuracy, reduce delays, strengthen access to diagnosis, and support more effective detection of drug-resistant TB. The Topic will also examine current gaps, implementation challenges, and opportunities for future improvement, including the integration of symptoms, radiological findings, laboratory data, and digital technologies into practical diagnostic pathways.We welcome submissions that address advances in laboratory and near-patient diagnosis of TB, with particular interest in studies assessing diagnostic performance, operational feasibility, clinical utility, cost-effectiveness, and real-world impact on case detection and the care cascade. Topics of interest include, but are not limited to:- Digital chest X-ray and computer-aided detection (CAD)- AI-powered acoustics, including cough-based screening applications and AI-enabled stethoscopes- AI-assisted smear microscopy and other tools for drug-resistant TB diagnostics- Pooling of biological samples to improve efficiency and testing coverage- Near-patient molecular diagnostics, including ultra-portable and minimal-infrastructure TB-NAAT platforms- Lateral flow and immunochromatographic tests in decentralised care pathways- Newer sequencing approaches, including targeted next-generation sequencing and CRISPR-based diagnostics- Performance, cost-effectiveness, and real-world impact of emerging diagnostic tools, including their implications for existing laboratory models and programme outcomesThis Research Topic will consider the following article types: Original Research, Review, Systematic Review, Meta-Analysis, Randomized Clinical Trial, Brief Research Report, Case Series, Community Case Study, Data Report, Methods, and Editorial.If you are interested in submitting or have any questions, please contact the journal email: tuberculosis@frontiersin.org. Topic editors: Sivakumar Shanmugam, Shaheed Vally Omar, Padmapriyadarsini Chandrasekaran Source: https://www.frontiersin.org/research-topics/79788/advances-in-tuberculosis-diagnosis-molecular-ai-enabled-and-decentralized-approaches