Federated Learning for Healthcare – Collaborative AI without Sharing Patient Data

Watch the video here

Artificial Intelligence and in particular Deep Learning has become an important technique for computer-aided medical applications. One of the common core requirements for learning a robust and accurate algorithm is the availability of an extensive and diverse data set. This, in turn, poses a major challenge for medical applications because patient data needs to be protected and cannot easily be shared. Federated learning addresses this problem by enabling to learn collaboratively across several institutions without explicitly sharing patient data.

This talk will give an introduction to the core concepts of Federated Learning (FL) and discusses the benefits, the unique considerations and challenges of implementing FL in the context of healthcare.

Nicola Rieke is a senior solution architect at NVIDIA for deep learning in healthcare and an active member of the medical imaging research community (e.g. Area Chair for MICCAI and IPCAI, organizer of various academic workshops). Throughout her studies and professional career, she has been working in the intersection of mathematics, medicine and computer science. In particular, she investigates real-time machine learning approaches for computer-assisted surgical interventions and federated learning for digital health. She holds a PhD from the Technical University of Munich, published various peer reviewed papers and was honored with the MICCAI Young Scientist Award.