用相似性加权聚合+差分隐私，保护隐私的同时提升脑肿瘤分割效果。
Similarity Weighted Aggregation with Global Differential Privacy for Federated Brain Lesion Segmentation
- 按样本相似度加权聚合模型更新，缓解数据异构问题。
- 在严格隐私预算下（ε=1），达到ET/TC/WT的Dice分数0.6357/0.5305/0.5274。
- 适合关注医疗图像隐私协作的科研与临床团队。
联邦学习可在不共享敏感数据的前提下实现多机构协同建模，特别适用于医学影像任务。然而，各机构间数据分布差异大且模型更新可能泄露信息仍是挑战。本文提出DP-SimAgg框架，结合相似性加权聚合与服务器端差分隐私机制：对参与方更新进行L2裁剪，计算基于相似度的聚合权重以缓解非独立同分布数据的影响，并在中心服务器注入校准高斯噪声，提供每轮隐私保障。该框架基于Intel OpenFL平台，在包含1251例多模态MRI的FeTS 2022数据集上评估。实验表明，即使在严格隐私预算（ε=1，累计ε_total=20，共20轮）下，仍保持良好分割性能，获得增强肿瘤（ET）、肿瘤核心（TC）、全肿瘤（WT）的Dice分数分别为0.6357、0.5305、0.5274；当放宽至ε=10时，性能接近无隐私基线，同时满足每轮(ε,δ)-DP要求。结果验证了该方法在医疗图像隐私协作中的潜力。
原文摘要 · Abstract (English)
Federated Learning (FL) enables collaborative training of machine learning models across multiple institutions without sharing sensitive data, making it particularly suitable for medical imaging applications. However, heterogeneous data distributions across institutions and potential information leakage through model updates remain important challenges. In this work, we propose DP-SimAgg, a privacy-preserving federated learning framework that integrates similarity-weighted aggregation with a server-side differential privacy mechanism. The proposed method applies L2 clipping to bound collaborator updates, computes similarity-based aggregation weights to mitigate the effects of non-IID data distributions, and injects calibrated Gaussian noise at the central server, providing per-round privacy guarantees under the assumed sensitivity bound. The framework is implemented using Intel's OpenFL platform and evaluated on the FeTS 2022 dataset consisting of 1251 multi-modal MRI scans for brain tumor segmentation. Experimental results demonstrate that DP-SimAgg maintains competitive segmentation performance while providing privacy protection. Under a strict per-round privacy budget (epsilon = 1, cumulative epsilon_total = 20 over 20 rounds), the method achieves Dice scores of 0.6357, 0.5305, and 0.5274 for the enhancing tumor (ET), tumor core (TC), and whole tumor (WT) regions, respectively. With a more relaxed per-round budget (epsilon = 10, cumulative epsilon_total = 200), performance approaches that of the non-private baseline while incorporating a central Gaussian mechanism with per-round (epsilon, delta)-DP accounting under the assumed sensitivity bound. These results highlight the potential of DP-SimAgg for enabling privacy-preserving collaborative learning in medical imaging applications.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。