Tuberculosis diagnosis using artificial intelligence: current trends and future prospects

TL;DR

This paper reviews how artificial intelligence can improve tuberculosis diagnosis by enhancing detection accuracy and sensitivity.

Contribution

The paper provides a comprehensive overview of machine learning-based models for automatic tuberculosis detection.

Findings

AI improves accuracy and sensitivity in detecting tuberculosis bacilli.

Chest radiography and pathology remain essential for tuberculosis diagnosis.

Low detection rates and false diagnoses persist due to challenges in microscopic identification.

Abstract

Tuberculosis, an infectious disease caused by Mycobacterium tuberculosis, poses a major global health challenge. Despite being largely controlled for several decades, tuberculosis has experienced a resurgence in recent years. China has the second highest incidence of tuberculosis globally, with a prevalence of 459 cases per 100,000 individuals aged 15 years old. Chest radiography and pathology are essential tools for its detection and diagnosis. However, the small size and low number of tubercle bacilli make detection and identification under a microscope challenging, often resulting in low detection rates and false diagnoses. Artificial intelligence (AI) has emerged as a promising tool to improve the accuracy and sensitivity of tuberculosis detection. This review provides a comprehensive overview of the literature on the use of machine learning-based models for the automatic detection of tuberculosis bacilli, emphasizing the advantages of integrating in tuberculosis diagnosis. Understanding the onset and progression of tuberculosis is crucial to developing effective strategies for its diagnosis, treatment, and prevention.

Figures