Currently, the biggest hurdle for tuberculosis screening using chest X-rays is the need for human experts to interpret the images. The project will demonstrate that automated, computerized X-ray interpretation TB screening (CAD4TB) is more effective, less costly, and accessible to millions of TB suspects at the bottom of the pyramid—the world’s poorest citizens. AIGHD’s main objective is to develop the decision analytic cost-effectiveness model to assess the cost-effectiveness of the CAD4TB technology.


Early case detection is the most promising strategy to reduce the enormous worldwide burden of tuberculosis (TB). Current tests such as the DNA-based GeneXpert test that has recently been endorsed by WHO and made available in many countries for access prices, are too costly and time-consuming to test every TB suspect. A much cheaper and faster solution is digital X-ray screening as an initial test to select who should undergo GeneXpert. But an X-ray requires human experts to interpret the image. The aim of the project is to demonstrate that the breakthrough technology CAD4TB, a computer software that automatically computes a score within a minute and can thus provide an immediate decision regarding which suspects should receive the more expensive and time-consuming GeneXpert test, is the perfect strategy for early case detection. We aim to do this by building a complete platform called CAD4TBCloud that can be used with any digital X-ray machine anywhere in the world and by providing a detailed cost-benefit analysis. The platform will be tested in Kigali, Rwanda, and Dhaka, Bangladesh to show that our solution is accessible to TB suspects at the bottom of the pyramid.


Decision analytic cost-effectiveness model and business case based on the model

AIGHD Research Lead

Prof. dr. Frank Cobelens


Rijksdienst voor ondernemend Nederland (RVO)