In short keynote presentations, Prof. Julia Schnabel (Director of Institute of Machine Learning in Biomedical Imaging) and Prof. Carsten Marr (Director of the Institute of AI for Health) explained how AI can help make diagnoses earlier and more accurate, simplify processes, and use resources more efficiently.
Julia Schnabel demonstrated how AI-based methods are transforming radiological imaging. With trained algorithms, typical patterns of healthy tissue structures can be identified, enabling faster and more precise evaluations. For example, MRI scans can be produced significantly faster while also improving in quality. This precision radiology opens new opportunities for early diagnosis – even before clinical symptoms appear – and enables significantly more efficient use of medical resources.
Carsten Marr illustrated how AI enhances pathological diagnostics at the single-cell level. For histological tissue sections, AI applications can generate findings – including disease classification and tumor typing. Reports that previously had to be created entirely manually can now be generated automatically and in a standardized manner. Marr emphasized that the success of such methods depends critically on a powerful infrastructure and access to high-quality data – only then can these advancements be reliably transferred into clinical care.
The joint discussion between researchers and members of the Bavarian State Parliament made it clear that AI applications cannot be viewed in isolation but rely on strong collaboration between science, clinical practice, data access, and computing power. Topics addressed included the long-term development of reliable funding structures, the importance of privacy-compliant research data, international competitiveness in high-performance computing, and the accelerated transfer of research results into practice – so that the benefits for patients can be realized quickly.
The Bavarian Health Cloud and the Hightech Agenda Bavaria were highlighted as key political initiatives that can help align innovation potential with data protection and societal value.
