Sachin Dudda Nagaraju

Joint segmentation and classification of bladder tumors can be achieved across institutions without centralizing sensitive patient data, leveraging federated learning for robust performance.

Achieving competitive medical image segmentation with just 10 annotated cases could revolutionize the efficiency of clinical workflows.

Simple global intensity nonlinear (GIN) augmentation enables surprisingly effective cross-modality generalization in federated medical image segmentation, outperforming complex alternatives and closing the gap with centralized training.