Distributed, Collaborative, and Federated Learning, and Affordable AI and Healthcare for Resource Diverse Global Health: Third MICCAI Workshop, DeCaF [Paperback]

This book constitutes the refereed proceedings of the Third MICCAI Workshop on Distributed, Collaborative, and Federated Learning, DeCaF 2022, and the Second MICCAI Workshop on Affordable AI and Healthcare, FAIR 2022, held in conjunction with MICCAI 2022, in Singapore in September 2022. FAIR 2022 was held as a hybrid event.

DeCaF 2022 accepted 14 papers from the 18 submissions received. The workshop aims at creating a scientific discussion focusing on the comparison, evaluation, and discussion of methodological advancement and practical ideas about machine learning applied to problems where data cannot be stored in centralized databases or where information privacy is a priority.

For FAIR 2022, 4 papers from 9 submissions were accepted for publication. The topics of the accepted submissions focus on deep ultrasound segmentation, portable OCT image quality enhancement, self-attention deep networks and knowledge distillation in low-regime setting.

Distributed, Collaborative, and Federated Learning.- Incremental Learning Meets Transfer Learning: Application to Multi-site Prostate MRI Segmentation .- FedAP: Adaptive Personalization in Federated Learning for Non-IID Data Data Stealing Attack on Medical Images: Is it Safe to Export Networks from Data Lakes? .- Data Stealing Attack on Medical Images: Is it Safe to Export Networks from Data Lakes?.- Can collaborative learning be private, robust and scalable?.- Split-U-Net: Preventing Data Leakage in Split Learning for Collaborative Multi-Modal Brain Tumor Segmentation.- Joint Multi Organ and Tumor Segmentation from Partial Labels using Federated Learning.- Fuh, Kensaku Mori, Weichung Wang, Holger R Roth GAN Latent Space Manipulation and Aggregation for Federated Learning in Medical Imaging.- A Specificity-Preserving Generative Model for Federated MRI Translation.- Content-Aware Differential Privacy with Conditional Invertible Neural Networks.- DeMed: A Novel and Efficienlc+