Center for Artificial Intelligence in Medicine (CAIM)

Events 2022

DAIM Lunch Talk

with Dr. Lisa Koch, Institute for Ophthalmic Research, University of Tübingen

(© Unsplash)

When:  18 October 2022, 12.00-13.10h
Where: ARTORG Center, Murtenstrasse 50, F 502 & online

CAIM is thrilled to host Lisa Koch, group leader "Machine Learning for Medical Diagnostics" for a talk at lunchtime. 

Please register prior to the hybrid event to reserve your seat or get your access link.
(Feel free to have your sandwich at the event.)

Towards Safe and Effective Medical Image Analysis Systems

Dr. Lisa Koch

University of Tübingen, Institute for Ophthalmic Research, Group "Machine Learning for Medical Diagnostics"


Over the last decades, medical imaging has become increasingly important in clinical practice as techniques for image acquisition and interpretation have matured. In particular, we have seen extraordinary progress in artificial intelligence (AI): Machine learning tools for medical image analysis, in particular if based on deep learning, are approaching human-level performance in controlled settings in various application areas such as automated detection and diagnosis systems or image segmentation. However, major hurdles still obstruct the wide adoption of AI in clinical practice. In real-world settings, in particular deep learning algorithms are famously brittle and are known to often fail silently and catastrophically. 

Since AI systems have the potential to cause harm to the patient, they are subject to regulations to ensure their safety and effectiveness. However, developing methods for demonstrating the safety and effectiveness of AI is still an active area of research. This talk provides an overview of the cornerstones of AI safety in medicine and introduces the pertinent work of Lisa Koch and her colleagues.


(photo by Silke Sommer)

Dr. Lisa Koch is a research group leader for machine learning in medical diagnostics at the University of Tübingen in Germany. Previously, she has been a postdoctoral researcher at ETH Zürich and did her doctorate at Imperial College London in the UK. Interspersed with academic research on machine learning in healthcare, she has gained several years of experience in industry.

She worked as a data scientist and later data science team lead at the Swiss medtech startup Ava, and a few years before that as a development engineer at Zühlke.

Lisa's research is inspired by her industry experience with the current challenges in deploying and approving artificial intelligence in the real world. Her focus lies in medical image analysis research that can ultimately benefit patient health in a safe and secure way, as well as in methods to facilitate certification and translation to real-world settings.