From Isolation to Collaboration: Federated Class-Heterogeneous Learning for Chest X-Ray Classification
This addresses a privacy-preserving challenge in medical imaging for healthcare applications, but it appears incremental as it builds on existing federated learning paradigms.
The paper tackles the problem of class-heterogeneity in federated learning for chest x-ray classification, where clients have different sets of classes, and proposes surgical aggregation, which outperforms current methods and shows better generalizability.
Federated learning (FL) is a promising paradigm to collaboratively train a global chest x-ray (CXR) classification model using distributed datasets while preserving patient privacy. A significant, yet relatively underexplored, challenge in FL is class-heterogeneity, where clients have different sets of classes. We propose surgical aggregation, a FL method that uses selective aggregation to collaboratively train a global model using distributed, class-heterogeneous datasets. Unlike other methods, our method does not rely on the assumption that clients share the same classes as other clients, know the classes of other clients, or have access to a fully annotated dataset. We evaluate surgical aggregation using class-heterogeneous CXR datasets across IID and non-IID settings. Our results show that our method outperforms current methods and has better generalizability.