提升医学影像分类模型抗干扰与泛化能力
Robust Training with Data Augmentation for Medical Imaging Classification
- 融合数据增强的鲁棒训练算法，统一优化抗干扰与泛化
- 在三种影像数据上均优于基线方法，保持高准确率
- 适合医疗AI部署，应对真实世界数据波动
深度神经网络在医学影像诊断中应用日益广泛。然而，这些模型极易受到对抗攻击和分布偏移的影响，降低诊断可靠性并削弱临床信任。本文提出一种结合数据增强的鲁棒训练算法（RTDA），以缓解医学图像分类中的此类脆弱性。我们在三种成像技术（乳腺钼靶、X光、超声）的数据集上，对比了RTDA与六种基线方法（包括对抗训练、单一或组合的数据增强）在对抗扰动和自然变异下的分类器鲁棒性。结果表明，RTDA在各项任务中均展现出更强的抗攻击能力与分布偏移下的泛化性能，同时保持了较高的原始准确率。
原文摘要 · Abstract (English)
Deep neural networks are increasingly being used to detect and diagnose medical conditions using medical imaging. Despite their utility, these models are highly vulnerable to adversarial attacks and distribution shifts, which can affect diagnostic reliability and undermine trust among healthcare professionals. In this study, we propose a robust training algorithm with data augmentation (RTDA) to mitigate these vulnerabilities in medical image classification. We benchmark classifier robustness against adversarial perturbations and natural variations of RTDA and six competing baseline techniques, including adversarial training and data augmentation approaches in isolation and combination, using experimental data sets with three different imaging technologies (mammograms, X-rays, and ultrasound). We demonstrate that RTDA achieves superior robustness against adversarial attacks and improved generalization performance in the presence of distribution shift in each image classification task while maintaining high clean accuracy.
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。