Federated Learning for Medical Imaging with Adaptive Client Sampling
Abstract
We present an adaptive client sampling strategy for federated training of medical image classifiers across 42 hospitals. The approach reduces straggler impact and achieves +2.8% AUROC over uniform sampling on chest X-ray anomaly detection.
Cite this article
Mitchell, C. & Lin, M. (2025). Federated Learning for Medical Imaging with Adaptive Client Sampling. Research Explorations in Global Knowledge & Technology (REGKT), 3 (7). Retrieved from https://regkt.com/article.php?id=131&slug=federated-learning-medical-imaging-adaptive-client-sampling