Decentralized Gossip Mutual Learning (GML) for brain tumor segmentation on multi-parametric MRI
This work addresses data privacy and efficiency challenges in medical imaging for hospitals, though it is incremental as it builds on existing federated learning methods.
The paper tackled the issues of server failures and suboptimal performance in Federated Learning for brain tumor segmentation by proposing Gossip Mutual Learning (GML), a decentralized framework using peer-to-peer communication and mutual learning, which achieved similar performance to FedAvg with only 25% communication overhead on the BraTS 2021 dataset.
Federated Learning (FL) enables collaborative model training among medical centers without sharing private data. However, traditional FL risks on server failures and suboptimal performance on local data due to the nature of centralized model aggregation. To address these issues, we present Gossip Mutual Learning (GML), a decentralized framework that uses Gossip Protocol for direct peer-to-peer communication. In addition, GML encourages each site to optimize its local model through mutual learning to account for data variations among different sites. For the task of tumor segmentation using 146 cases from four clinical sites in BraTS 2021 dataset, we demonstrated GML outperformed local models and achieved similar performance as FedAvg with only 25% communication overhead.