Machine learning-based algorithms for early detection of tuberculosis: A scoping review

Abstract

Tuberculosis (TB) remains a major global health challenge, particularly in low- and middle-income countries. Early and accurate detection is crucial for effective disease control, timely treatment initiation, and improved patient outcomes. Conventional diagnostic methods, such as smear microscopy and culture, are limited by suboptimal sensitivity, variable specificity, and long turnaround times. Machine learning (ML) has emerged as a promising tool to enhance diagnostic accuracy, especially through biomarker-based approaches. Techniques such as radiomics and deep learning (DL) enable early detection by identifying disease patterns in biomarker data, facilitating rapid and precise diagnosis. The aim of this study was to explore the role of ML in biomarker-based TB detection, emphasizing its potential in early diagnosis and improved screening strategies. A comprehensive review of studies on ML applications in biomarker-based early detection and diagnosis of TB was conducted using PubMed, Scopus, Web of Science, and Epistemonikos databases. Studies published up to February 3, 2025, were analyzed to evaluate the effectiveness of ML models in TB detection. ML-based diagnostic models, including those incorporating radiomics and DL, have demonstrated promising results in TB early detection and diagnosis. Studies indicate that ML techniques can enhance sensitivity and specificity in biomarker-based TB screening, enabling earlier intervention. The study highlights key challenges, such as data variability, model generalizability, and the need for standardized validation methods to ensure clinical applicability. ML offers a valuable approach for biomarker-based TB early detection and diagnosis, addressing limitations in conventional methods. While the current findings support its potential, further research is needed to optimize model performance, enhance reproducibility, and facilitate clinical integration for more effective TB control strategies.