用生成模型增广乳腺超声图像，提升联邦学习检测效果
Federated Breast Cancer Detection Enhanced by Synthetic Ultrasound Image Augmentation
- 结合GAN与扩散模型生成合成超声图增强数据
- 合理使用合成数据使AUC最高提升0.0145
- 适合医疗联邦学习、数据稀缺场景研究者
联邦学习可在不共享敏感患者数据的前提下实现多机构协同训练深度学习模型。然而，其性能常受限于小样本数据和非独立同分布的数据分布，影响模型泛化能力。本文提出一种基于生成模型的数据增强框架，用于乳腺超声分类任务。该框架利用深度卷积生成对抗网络（DC-GAN）和类别条件去噪扩散概率模型生成合成图像。在三个公开数据集（BUSI、BUS-BRA 和 UDIAT）上的实验表明，适当数量的合成图像可将 FedAvg 的平均 AUC 从 0.9206 提升至 0.9362，FedProx 从 0.9429 提升至 0.9574。同时发现，过度使用合成数据会降低性能，凸显了真实与合成样本平衡的重要性。结果表明，基于生成模型的增强策略能有效提升联邦乳腺超声图像分类性能。
原文摘要 · Abstract (English)
Federated learning enables collaborative training of deep learning models across institutions without sharing sensitive patient data. However, its performance is often limited by small datasets and non-independent, identically distributed data, which can impair model generalization. In this work, we propose a generative model-based data augmentation framework for breast ultrasound classification. It leverages synthetic images generated by deep convolutional generative adversarial networks and a class-conditioned denoising diffusion probabilistic model. Experiments on three publicly available datasets (BUSI, BUS-BRA, and UDIAT) demonstrated that incorporating a suitable number of synthetic images improved average AUC from 0.9206 to 0.9362 for FedAvg and from 0.9429 to 0.9574 for FedProx. Furthermore, we noticed that excessive use of synthetic data reduced performance. This highlights the importance of balancing real and synthetic samples. Our results underscore the potential of generative model-based augmentation to enhance federated breast ultrasound image classification.
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。