From partly to fully automated healthcare, and beyond

by Prof. Bram van Ginneken (Radboud UMC Nijmegen)

It took more than half a century, but now computer vision – the dream of building machines that can see – has been solved. Our focus should shift from research to implementation. In this talk, I will share my experiences in bringing various medical image analysis algorithms to the product level and have an impact on healthcare. I will argue that the greatest gains lie not in making healthcare better and even more expensive, but in making it reasonably good and broadly affordable. Beyond this, I will place these efforts in a larger context. Our society is under great pressure from an aging population and environmental catastrophe. Paradoxically, the technology seen by many as humanity’s greatest danger – AI – seems to be the only way to navigate these crises. I expect we still have a fighting chance to achieve the best possible outcome, summarized by the great thinker Aaron Bastani as “fully automated luxury communism”.

Bram van Ginneken was born in 1970 in Nuenen. He studied Physics at Eindhoven University of Technology and Utrecht University. In 1996, he started working on medical image analysis, and started to climb the academic ladder as PhD student, assistant professor, associate professor and full professor. He set up and led the Diagnostic Image Analysis Group at Radboud UMC in Nijmegen from 2010 onwards. In 2023, he decided to step down to focus on a new company, Plain Medical, to improve the efficiency of radiology reporting. Previously, he developed products with Delft Imaging for tuberculosis screening, with MeVis Medical Solutions for lung cancer screening, and with Thirona (a company he co-founded) for chest CT and retinal image analysis. Additionally, he pioneered the concept of challenges in medical image analysis and created grand-challenge.org.

Other keynotes at our Winter Meeting 2025:

• Computational imaging methods on artworks from the Rijksmuseum collection by Džemila Šero
• Building stuff that works by Tom Koopen