A Survey of Incremental Transfer Learning: Combining Peer-to-Peer Federated Learning and Domain Incremental Learning for Multicenter Collaboration

TL;DR

This paper surveys incremental transfer learning methods that combine federated learning and domain incremental learning to enable privacy-preserving multicenter collaboration in medical imaging, analyzing various factors affecting performance.

Contribution

It adapts a domain/task incremental learning framework for transfer learning and provides a comprehensive survey of regularization-based continual learning methods for multicenter collaboration.

Findings

Data heterogeneity impacts model performance.

Classifier head setting influences transfer effectiveness.

Model initialization and center order affect learning outcomes.

Abstract

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.