Keynote by Prof. Dr. Lisa Koch on Explainable AI in medical imaging: promises, a reality check and ways forward, 20 May 2025
Tuesday, 20 May 2025, 4.15 pm CET, ZOOM
About Lisa Koch
Prof. Lisa Koch is an assistant professor for data science in diabetes care at the University of Bern, the University of Bern Clinic for Diabetology, Endocrinology, Nutritional Medicine and Metabolism (UDEM) and the Diabetes Center Berne (DCB). Prof. Koch’s goal is to develop certifiably safe, reliable and effective machine learning tools for improving patient care. She has a background in academic research as well as developing data science products for medical devices. Prof. Lisa Koch holds undergraduate degrees in electrical (BSc) and biomedical engineering (MSc) from ETH Zürich and a PhD in Computer Science from Imperial College London, UK. After a post-doc at ETH Zürich she joined the Swiss wearable medical device startup Ava, where she eventually became the data science team lead. In this position, she came to appreciate the need for demonstrably safe machine learning in healthcare. In 2021, she returned to academic research to pursue research on this topic as a group leader for machine learning in medical diagnostics at the Hertie Institute for AI in Brain Health at the University of Tübingen, Germany.
About the Lecture
Recent breakthroughs in medical AI are largely based on black-box deep neural networks. While such systems have achieved remarkable performance, even surpassing clinical experts in some applications, black-box AI systems can fail silently and unexpectedly. Explainable AI techniques, which aim to open the black-box and make deep learning models transparent, are often considered a solution to establishing trust and accountability. However, explainability techniques are often flawed themselves, and it is unclear what practical value they can actually provide for improving patient care. In this lecture, Prof. Koch will provide an overview of the current state-of-the-art of explainable AI in medicine including their limitations and recent innovations. She will then describe potential use cases for explainable AI in clinical workflows, and outline alternative ways of establishing trust in medical AI systems beyond explainable AI.
We are pleased to invite you to join this guest lecture, which is part of our CDHI Lecture Series Digital Health Forum. Registration is not required. Please be aware that we will be recording this guest lecture and will be making it available in our teaching library. If you have any questions, please contact Victoria Brügger (victoria.bruegger@unisg.ch) prior to the start of the guest lecture.
Prof. Dr. Tobias Kowatsch, Associate Professor for Digital Health Interventions, Institute for Implementation Science in Health Care, University of Zurich; Director, School of Medicine, University of St.Gallen (HSG); Scientific Director, Centre for Digital Health Interventions (CDHI), ETH Zurich & HSG; Principal Investigator, Future Health Technologies programme, Campus for Research Excellence and Technological Enterprise (CREATE), Singapore-ETH Centre, Singapore
Prof. Dr. Elgar Fleisch, Professor of Information Management, ETH Zürich; Professor of Technology Management, University of St.Gallen; Advisory Board Member, CDHI, ETH Zürich & University of St.Gallen; Principal Investigator, Future Health Technologies programme, Campus for Research Excellence and Technological Enterprise (CREATE), Singapore-ETH Centre, Singapore
Prof. Dr. Florian von Wangenheim, Professor of Technology Marketing, ETH Zurich & Advisory Board Member, CDHI; Principal Investigator, Future Health Technologies programme, Campus for Research Excellence and Technological Enterprise (CREATE), Singapore-ETH Centre, Singapore
