Artificial intelligence systems are already used in gastrointestinal endoscopy to support doctors in the early detection of various conditions. How well these systems perform, however, depends heavily on the quality of the endoscopic procedure itself. An AI system can only reliably recognise abnormalities if the images it is trained on and works with are of sufficient quality.

With the Veni grant, De Groof is developing three computer-aided quality systems, known as CAQ systems, designed to monitor and improve this underlying quality. The first system monitors quality during a colonoscopy and provides doctors with feedback they can act on immediately. The second system assesses the quality of images of polyps. The third system will be clinically validated for the surveillance of Barrett’s oesophagus. Together, the three systems aim to standardise the quality of endoscopic care and increase the reliability of existing AI systems in practice.

The research is one of fifteen Veni projects starting this year at Amsterdam UMC. The Veni grant is awarded annually to promising researchers who have recently completed their doctorates and provides up to 320,000 euros, giving recipients three years to pursue innovative research.