Abstract:Federated learning (FL) enables multiple sites to collaboratively train powerful deep models without compromising data privacy and security. The statistical heterogeneity (e.g., non-IID data and domain shifts) is a primary obstacle in FL, impairing the generalization performance of the global model. Weakly supervised segmentation, which uses sparsely-grained (i.e., point-, bounding box-, scribble-, block-wise) supervision, is increasingly being paid attention to due to its great potential of reducing annotation costs. However, there may exist label heterogeneity, i.e., different annotation forms across sites. In this paper, we propose a novel personalized FL framework for medical image segmentation, named FedICRA, which uniformly leverages heterogeneous weak supervision via adaptIve Contrastive Representation and Aggregation. Concretely, to facilitate personalized modeling and to avoid confusion, a channel selection based site contrastive representation module is employed to adaptively cluster intra-site embeddings and separate inter-site ones. To effectively integrate the common knowledge from the global model with the unique knowledge from each local model, an adaptive aggregation module is applied for updating and initializing local models at the element level. Additionally, a weakly supervised objective function that leverages a multiscale tree energy loss and a gated CRF loss is employed to generate more precise pseudo-labels and further boost the segmentation performance. Through extensive experiments on two distinct medical image segmentation tasks of different modalities, the proposed FedICRA demonstrates overwhelming performance over other state-of-the-art personalized FL methods. Its performance even approaches that of fully supervised training on centralized data. Our code and data are available at <a class="link-external link-https" href="https://github.com/llmir/FedICRA" rel="external noopener nofollow">this https URL</a>.

FedA3I: Annotation Quality-Aware Aggregation for Federated Medical Image Segmentation against Heterogeneous Annotation Noise

FedIA: Federated Medical Image Segmentation with Heterogeneous Annotation Completeness

FedGrav: An Adaptive Federated Aggregation Algorithm for Multi-institutional Medical Image Segmentation

Unifying and Personalizing Weakly-supervised Federated Medical Image Segmentation via Adaptive Representation and Aggregation

FedDM: Federated Weakly Supervised Segmentation via Annotation Calibration and Gradient De-conflicting

FedATA: Adaptive Attention Aggregation for Federated Self-supervised Medical Image Segmentation

Medical federated learning with joint graph purification for noisy label learning

Personalized federated learning for abdominal multi‐organ segmentation based on frequency domain aggregation

FedSODA: Federated Cross-assessment and Dynamic Aggregation for Histopathology Segmentation

FedLPPA: Learning Personalized Prompt and Aggregation for Federated Weakly-supervised Medical Image Segmentation

FEDKA: Federated Knowledge Augmentation for Multi-Center Medical Image Segmentation on Non-Iid Data

Federated Data Quality Assessment Approach: Robust Learning With Mixed Label Noise

Quality-Adaptive Split-Federated Learning for Segmenting Medical Images with Inaccurate Annotations

Vicinal Feature Statistics Augmentation for Federated 3D Medical Volume Segmentation

Reliable Mutual Distillation for Medical Image Segmentation under Imperfect Annotations

Federated Model Aggregation via Self-Supervised Priors for Highly Imbalanced Medical Image Classification

FedAA: Using Non-sensitive Modalities to Improve Federated Learning while Preserving Image Privacy

Collaboratively Learning Federated Models from Noisy Decentralized Data

Federated Multi-organ Segmentation with Inconsistent Labels

Learning Unlabeled Clients Divergence via Anchor Model Aggregation for Federated Semi-supervised Learning

Learning Unlabeled Clients Divergence for Federated Semi-Supervised Learning via Anchor Model Aggregation