Background
Tuberculosis (TB) is an infectious disease caused by Mycobacterium tuberculosis and remains one of the leading causes of infectious disease-related deaths worldwide. Primarily affecting the lungs, TB is transmitted through airborne particles, making early detection and diagnosis essential to limit disease progression and transmission.
Chest X-ray (CXR) is widely used for TB screening, allowing the identification of pulmonary abnormalities such as nodules, infiltrates, and cavitary lesions. Computed tomography (CT) provides more detailed anatomical visualisation, enabling improved characterisation of these abnormalities. Additionally, PET-CT combines anatomical and functional imaging to assess metabolic activity associated with infection and inflammation, offering valuable insights into disease activity and treatment response.
Together, these imaging modalities provide complementary information for the detection, characterisation, and monitoring of tuberculosis.
Aim
The XACT-19 project is a collaboration between Radboudumc, Delft Imaging, and the University of Cape Town Lung Institute, South Africa.
The project aims to investigate the reliability of chest X-ray (CXR) for tuberculosis (TB) detection using computer-aided detection software (CAD4TB). Patients with suspected TB are included in the study and undergo CXR, CT, and PET imaging at the participating hospital in Cape Town, South Africa. PET imaging provides additional information on metabolic activity associated with TB and is used to support the evaluation of CXR findings and CAD4TB performance. Furthermore, the project explores tools for the (automated) analysis and characterisation of TB-related abnormalities in medical images.
Funding
The XACT-19 project is funded by the European and Developing Countries Clinical Trials Partnership (EDCTP grant RIA2020S-3295)





