AI in healthcare

AI can now create “virtual patients” to help us test new treatments without risk—but how do we know we can trust them? In this talk, I’ll explore how synthetic data and digital twins are being used to improve procedures like Transcatheter Aortic Valve Implantation (TAVI), and how we validate these models to ensure they truly reflect real patients.

Your future eye test might do more than check your glasses. Advances in artificial intelligence show that the eye carries subtle signals about overall health and ageing. Using routine scans, AI can detect patterns linked to stroke risk, dementia, and heart disease, often before symptoms appear. But how reliable are these predictions, and what should we do with them? This talk explores the science behind these signals, recent advances, and the challenges of bringing this technology into everyday care.

The use of AI and machine learning in medical devices is increasing year on year. While some highlight risks such as bias, lack of transparency, and uncertainty, others see significant potential to improve patient care. Are these concerns justified, or part of a broader fear of new technology? This talk explores the real risks, the misconceptions, and what it takes to make AI safe and trustworthy in healthcare.