A Survey of Incremental Transfer Learning: Combining Peer-to-Peer Federated Learning and Domain Incremental Learning for Multicenter Collaboration
This work addresses data privacy issues in multicenter clinical collaborations, but it is incremental as it adapts existing frameworks and surveys methods.
The paper tackles the problem of developing high-performance deep learning models from multicenter clinical data while respecting privacy constraints, by proposing an incremental transfer learning framework that combines peer-to-peer federated learning and domain incremental learning, and it investigates factors like data heterogeneity and model initialization to preserve performance.
Due to data privacy constraints, data sharing among multiple clinical centers is restricted, which impedes the development of high performance deep learning models from multicenter collaboration. Naive weight transfer methods share intermediate model weights without raw data and hence can bypass data privacy restrictions. However, performance drops are typically observed when the model is transferred from one center to the next because of the forgetting problem. Incremental transfer learning, which combines peer-to-peer federated learning and domain incremental learning, can overcome the data privacy issue and meanwhile preserve model performance by using continual learning techniques. In this work, a conventional domain/task incremental learning framework is adapted for incremental transfer learning. A comprehensive survey on the efficacy of different regularization-based continual learning methods for multicenter collaboration is performed. The influences of data heterogeneity, classifier head setting, network optimizer, model initialization, center order, and weight transfer type have been investigated thoroughly. Our framework is publicly accessible to the research community for further development.