Encrypted federated learning for secure decentralized collaboration in cancer image analysis

Abstract

Abstract Artificial Intelligence (AI) has a multitude of applications in cancer research and oncology. However, the training of AI systems is impeded by the limited availability of large datasets due to data protection requirements and other regulatory obstacles. Federated and swarm learning represent possible solutions to this problem by collaboratively training AI models while avoiding data transfer. However, in these decentralized methods, weight updates are still transferred to the aggregation server for merging the models. This leaves the possibility for a breach of data privacy, for example by model inversion or membership inference attacks by untrusted servers. Homomorphically encrypted federated learning (HEFL) is a solution to this problem because only encrypted weights are transferred, and model updates are performed in the encrypted space. Here, we demonstrate the first successful implementation of HEFL in a range of clinically relevant tasks in cancer image analysis on multicentric datasets in radiology and histopathology. We show that HEFL enables the training of AI models which outperform locally trained models and perform on par with models which are centrally trained. In the future, HEFL can enable multiple institutions to co-train AI models without forsaking data governance and without ever transmitting any decryptable data to untrusted servers. One Sentence Summary Federated learning with homomorphic encryption enables multiple parties to securely co-train artificial intelligence models in pathology and radiology, reaching state-of-the-art performance with privacy guarantees.

Ähnliche Arbeiten