arXiv 2605.005782026-05-19cs.CV

通过高斯混合特征对齐和课程整合实现全切片图像的联邦蒸馏

Federated Distillation for Whole Slide Image via Gaussian-Mixture Feature Alignment and Curriculum Integration

School of Computer Science, Peking University, Beijing, China（北京大学计算机科学系）；Center for Health Informatics, Australian Institute of Health Innovation, Macquarie University, Sydney, NSW 2113, Australia（健康信息学中心，澳大利亚健康创新研究所，麦考利大学，悉尼，NSW 2113，澳大利亚）；School of Data Science, University of Virginia, Charlottesville, VA, USA（数据科学学院，弗吉尼亚大学，夏洛特维尔，VA，美国）

Luru Jing, Cong Cong, Yanyuan Chen, Yongzhi Cao

备注：: Accepted by ICML 2026, Camera-Ready version updated

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出FedHD框架，通过高斯混合特征对齐和课程整合策略，在联邦学习中实现全切片图像分析，通过本地生成的语义丰富合成特征表示提升模型性能，同时保持诊断多样性。

摘要：

联邦学习（FL）提供了一个有前景的框架，用于通过跨机构进行模型训练来实现协作数字病理学。然而，现实部署面临异质性问题，源于不同机构中多样化的多实例学习（MIL）架构和异构特征提取器。我们提出FedHD，一种新的FL框架，通过针对WSI分析进行本地高斯混合特征对齐。不同于交换模型参数，每个客户端独立地蒸馏语义丰富的合成特征表示，这些表示与真实WSI的分布对齐。为保持诊断多样性，FedHD采用一对一蒸馏策略，为每个真实切片生成一个合成对应物，以避免过度压缩。在联邦过程中，采用基于课程的整合策略，一旦性能达到平台期，逐步将跨站点的合成特征整合到本地训练中。此外，一个可选的解释模块从合成嵌入中重建伪块，提高透明度。FedHD是架构无关的、隐私保护的，并支持在不同机构之间进行个性化但协作的训练。在TCGA-IDH、CAMELYON16和CAMELYON17上的实验表明，FedHD在联邦和蒸馏基线中表现一致优于最先进的方法。

英文摘要：

Federated learning (FL) offers a promising framework for collaborative digital pathology by enabling model training across institutions. However, real-world deployments face heterogeneity arising from diverse multiple instance learning (MIL) architectures and heterogeneous feature extractors across institutions. We propose FedHD, a novel FL framework that performs local Gaussian-mixture feature alignment tailored for WSI analysis. Instead of exchanging model parameters, each client independently distills semantically rich synthetic feature representations aligned with the distribution of real WSIs. To preserve diagnostic diversity, FedHD adopts a one-to-one distillation strategy, generating a synthetic counterpart for each real slide to avoid over-compression. During federation, a curriculum-based integration strategy progressively incorporates cross-site synthetic features into local training once performance plateaus. Furthermore, an optional interpretation module reconstructs pseudo-patches from synthetic embeddings, enhancing transparency. FedHD is architecture-agnostic, privacy-preserving, and supports personalized yet collaborative training across diverse institutions. Experiments on TCGA-IDH, CAMELYON16, and CAMELYON17 show that FedHD consistently outperforms state-of-the-art federated and distillation baselines.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。