College of Chemistry and Life Science, Beijing University of Technology, Beijing, People's Republic of China.
Beijing International Science and Technology Cooperation Base for Intelligent Physiological Measurement and Clinical Transformation, Beijing, People's Republic of China.
Phys Med Biol. 2024 May 8;69(10). doi: 10.1088/1361-6560/ad4083.
Deep neural networks (DNNs) have been widely applied in medical image classification and achieve remarkable classification performance. These achievements heavily depend on large-scale accurately annotated training data. However, label noise is inevitably introduced in the medical image annotation, as the labeling process heavily relies on the expertise and experience of annotators. Meanwhile, DNNs suffer from overfitting noisy labels, degrading the performance of models. Therefore, in this work, we innovatively devise a noise-robust training approach to mitigate the adverse effects of noisy labels in medical image classification. Specifically, we incorporate contrastive learning and intra-group mixup attention strategies into vanilla supervised learning. The contrastive learning for feature extractor helps to enhance visual representation of DNNs. The intra-group mixup attention module constructs groups and assigns self-attention weights for group-wise samples, and subsequently interpolates massive noisy-suppressed samples through weighted mixup operation. We conduct comparative experiments on both synthetic and real-world noisy medical datasets under various noise levels. Rigorous experiments validate that our noise-robust method with contrastive learning and mixup attention can effectively handle with label noise, and is superior to state-of-the-art methods. An ablation study also shows that both components contribute to boost model performance. The proposed method demonstrates its capability of curb label noise and has certain potential toward real-world clinic applications.
深度神经网络（DNN）在医学图像分类中得到了广泛的应用，并取得了显著的分类性能。这些成果在很大程度上依赖于大规模的准确标注训练数据。然而，医学图像标注过程中不可避免地会引入标签噪声，因为标注过程严重依赖标注者的专业知识和经验。同时，DNN 容易受到噪声标签的过度拟合的影响，从而降低模型的性能。因此，在这项工作中，我们创新性地设计了一种抗噪训练方法，以减轻医学图像分类中噪声标签的不利影响。具体来说，我们将对比学习和组内混合注意力策略结合到了传统的监督学习中。特征提取器的对比学习有助于增强 DNN 的视觉表示。组内混合注意力模块构建了组，并为组内样本分配了自注意力权重，然后通过加权混合操作对大量噪声抑制样本进行插值。我们在不同噪声水平下的合成和真实世界的噪声医学数据集上进行了对比实验。严格的实验验证了我们的具有对比学习和混合注意力的抗噪方法可以有效地处理标签噪声，并优于最先进的方法。消融研究还表明，两个组成部分都有助于提高模型性能。所提出的方法证明了其抑制标签噪声的能力，并在实际临床应用中具有一定的潜力。
