Stellenbosch 肆客足球 (SU) researchers are playing a leading role in a global trial investigating the use of artificial intelligence (AI) to improve tuberculosis (TB) diagnosis. The project aims to develop and test an algorithm that will enable healthcare workers at primary care facilities to detect likely TB cases using a handheld ultrasound device and smartphone.
“TB remains the world's deadliest infectious disease, yet it is massively underdiagnosed," explains Prof Grant Theron, professor in Clinical Mycobacteriology and Epidemiology at SU and the trial coordinator. “A major challenge is that we often test the wrong people at the wrong time. Many patients undergo unnecessary testing, while others who desperately need it never receive proper screening. There's an urgent need for accessible, affordable, and scalable diagnostic tools for TB triage."
The project, titled 'Computer assisted diagnosis with lung ultrasound for community based pulmonary tuberculosis triage in Benin, Mali and South Africa' (CAD LUS4TB), involves a consortium of 10 health and research institutions across Africa and Europe. The European Union's Global Health EDCTP3 Joint Undertakings has provided �10 million (over R200 million) in funding.
The study will include 3 000 adult patients to investigate ultrasound-led TB recognition using AI in TB triage and management. The goal is to improve access to TB screening that can rule out TB disease among symptomatic adult patients at the primary healthcare level.
"Point-of-care lung ultrasound employs sensitive, handheld imaging devices capable of detecting body abnormalities, including those characteristic of TB," Theron explains. "Previously, this technology was limited by the need for specialised expertise to interpret images. However, AI now offers unprecedented opportunities to automate image classification, allowing minimally trained health workers to quickly and easily determine which patients require further testing. CAD LUS4TB therefore introduces a much-needed, specimen-free diagnostic test in the fight against TB."
In collaboration with European partners, SU will also develop and validate the machine learning algorithms with the involvement of Prof Thomas Niesler's Digital Signal Processing group in SU's Faculty of Engineering. Researchers will develop the novel algorithm to be compatible with portable ultrasound devices that connect to smartphones. The technology will automatically assess ultrasound images for TB indicators and will be packaged into a user-friendly mobile application for widespread deployment.
The project begins on September 1, 2025, under the co-leadership of Dr Veronique Suttels from The Swiss Federal Technology Institute of Lausanne Laboratory for Intelligent Global Health and Humanitarian Technologies, and Prof Ablo Prudence Wachinou from the National Teaching Centre for Pneumology & Tuberculosis in Benin.
The CAD LUS4TB consortium focuses on generating population-specific evidence and advocating for the integration of computer-assisted diagnosis (CAD) using AI to support the implementation of lung ultrasound in healthcare policy.
Caption : A doct?or at the national teaching hospital for pneumology & tuberculosis in Cotonou, Benin performs a tablet-connected ultrasound on a symptomatic patient. CAD LUS4TB investigates AI-assisted image analysis to enable early detection of tuberculosis.
