ARTIFICIAL INTELLIGENCE IN TUBERCULOSIS DETECTION: A REVIEW OF QXR AND ITS CLINICAL IMPACT

Tuberculosis (TB) continues to be a major public health concern, especially in developing countries where early diagnosis and timely treatment are critical for controlling its spread. Recent advances in Artificial Intelligence (AI) have opened new possibilities in medical diagnostics. One such innovation is qXR, an AI-powered tool designed to detect TB by analyzing chest X-rays quickly and accurately. This review paper focuses on the performance, accuracy, and clinical usefulness of qXR in TB detection. qXR has been trained using large datasets and is capable of scanning chest X-rays without human assistance. In one study, it analysed 1,675 chest X-rays and identified potential TB cases with 87% accuracy. These results were then verified by expert radiologists, confirming the tool’s reliability. Additionally, qXR was able to detect 17% more cases that could have been missed during routine examination. This highlights its role in improving early diagnosis. The tool complies with World Health Organization (WHO) guidelines and offers a faster and more efficient alternative to traditional radiography. By providing rapid screening, qXR helps medical professionals start treatment earlier, which can significantly improve patient outcomes. While AI cannot replace the role of radiologists, it can assist them by reducing workload and enhancing diagnostic accuracy. This review concludes that qXR is a valuable tool for preliminary TB screening and has the potential to transform TB detection, particularly in resource-limited settings.