DIAGNOSTIC ACCURACY AND EFFECTIVENESS OF COMPUTER-AIDED DETECTION (CAD/AI) SYSTEMS IN THE ANALYSIS OF CHEST X-RAY IMAGES FOR THE EARLY DETECTION OF PULMONARY TUBERCULOSIS

This article examines the role and effectiveness of Computer-Aided Detection (CAD) and Artificial Intelligence (AI) algorithms in the early detection and differential diagnosis of pulmonary tuberculosis. Based on World Health Organization (WHO) recommendations and a review of contemporary international scientific literature, the diagnostic sensitivity and specificity of AI systems based on convolutional neural networks (CNNs) are analyzed. Particular attention is paid to the influence of human factors on the detection of focal and infiltrative abnormalities in digital chest X-ray images (dCXR), as well as to the practical and economic advantages of integrating AI/CAD systems into screening and triage workflows.