By Matthew Brady 
In a Lancet Digital Health-published study of 12 AI-powered computer-aided detection (CAD) products, led by digital health specialist Dr. Zhi Zhen Qin from Heidelberg University Hospital in Germany and a research team from Stop TB Partnership, Lunit INSIGHT CXR showed the best overall performance for TB detection.
Lunit INSIGHT CXR, a AI-powered chest X-ray analysis solution from Korean healthcare company Lunit, achieved an AUC of 0.902 using digital chest X-ray images and metadata individuals who participated in a South African TB prevalence survey.
The solution achieved a specificity of 67.7%, the closest among the 12 CAD products evaluated to the WHO target of 70% specificity for TB classification in individuals over 15, when set to 90% sensitivity.
According to Lunit CEO Brandon Soh, the solution’s capacity to maintain high sensitivity across diverse populations and diagnostic thresholds is “crucial in resource-limited settings where each undetected case can can significant repercussions,” highlighting the continued impact of TB in developing countries worldwide.