Deepfakes in Medicine – Progress or a Risk to Patients? – MDR Wissen
Interview with Prof. Dr Thomas Neumuth, PI in the KIMed Network
News | 10.05.2026
AI-generated images are no longer just an issue on social media – they are now also being used in medicine. A recent study published in the journal Radiology shows that neither experienced radiologists nor specialised AI systems can reliably distinguish artificially generated X-ray images from real ones. A worrying finding?
In an interview with the German public broadcaster MDR, Prof. Dr Thomas Neumuth (Leipzig University) and Prof. Dr Felix Nensa (University of Duisburg-Essen) discuss and assess this development.
Professor Neumuth’s conclusion: “Any risk that exists can also be addressed through appropriate countermeasures.”
In the interview, Professor Neumuth also discusses:
- Possible measures for the safe handling of synthetic images in medicine
- The potential of synthetic data in medicine, for example for testing new systems, supplementing missing data or training healthcare professionals
- The importance of AI literacy and media literacy in everyday medical practice
Prof. Dr Thomas Neumuth is Technical Director (CTO) at the Innovation Center for Computer Assisted Surgery (ICCAS) at Leipzig University and represents the Leipzig site within the KIMed network as a Principal Investigator.
