Federated learning (FL) enables privacy-preserving collaborative model training without direct data sharing. Model-heterogeneous FL (MHFL) enables clients to train personalized models with heterogeneous architectures, but existing methods mainly rely on centralized aggregation or require partially identical architectures, limiting scalability and efficiency. Current peer-to-peer (P2P) FL frameworks, though removing server dependence, have not been adapted to heterogeneous models and suffer from model drift and knowledge dilution. To address these challenges, we propose FedSKD, a novel P2P MHFL framework for medical image classification that facilitates direct knowledge exchange through round-robin model circulation, eliminating the need for centralized aggregation while allowing fully heterogeneous model architectures across clients. FedSKD's key innovation lies in multidimensional similarity knowledge distillation (SKD), which enables bidirectional cross-client knowledge transfer at batch, pixel/voxel, and region levels for heterogeneous models in FL. This approach mitigates catastrophic forgetting and model drift through progressive reinforcement and distribution alignment while preserving model heterogeneity. Extensive evaluations on fMRI-based autism spectrum disorder (ASD) diagnosis and skin lesion classification demonstrate that FedSKD outperforms state-of-the-art heterogeneous and homogeneous FL baselines, achieving superior personalization and cross-institutional generalization. These findings underscore FedSKD's potential as a scalable and robust solution for real-world medical FL.
Weng et al. (2026) studied this question.