Deep federated learning in healthcare
Learning to distill and share knowledge among AI agents
Federated learning (FL) has been recently introduced to enable training deep learning (DL) models or AI agents without sharing the data. In other words, AI agents at local hubs, e.g. hospitals, are trained on their own data and share only the trained parameters with a centralized AI model or other AI agents. Leveraging such a massive amount of data in a privacy-preserved fashion would have a great impact on outbreak detection, drug development, and other healthcare services.
Yet, principal challenges, to overcome, concern the nature of medical data, namely data heterogeneity; severe class-imbalance, few amounts of annotated data, inter-/intra-scanners variability (domain shift), inter-/intra- observer variability (noisy annotations); system heterogeneity, and explainability and robustness.
The goal of this Helmholtz AI young investigator group is to develop innovative deep FL algorithms that can distill and share the knowledge among AI agents in a robust and privacy-preserved fashion.