Personalized and privacy-preserving federated heterogeneous medical image analysis with PPPML-HMI

Juexiao Zhou, Longxi Zhou, Di Wang, Xiaopeng Xu, Haoyang Li, Yuetan Chu, Wenkai Han, Xin Gao

Abstract: Highlights•A robust and generalizable privacy-preserving personalized federal learning method for heterogeneous medical image analysis.•A novel decentralized cyclic secure aggregation mechanism with homomorphic encryption.•PPPML-HMI has demonstrated high performance and robustness in two real-life cases.