IEEE J Biomed Health Inform. 2021 Apr;25(4):1163-1172. doi: 10.1109/JBHI.2020.3021558. Epub 2021 Apr 6.
In recent years, deep learning methods have received more attention in epithelial-stroma (ES) classification tasks. Traditional deep learning methods assume that the training and test data have the same distribution, an assumption that is seldom satisfied in complex imaging procedures. Unsupervised domain adaptation (UDA) transfers knowledge from a labelled source domain to a completely unlabeled target domain, and is more suitable for ES classification tasks to avoid tedious annotation. However, existing UDA methods for this task ignore the semantic alignment across domains. In this paper, we propose a Curriculum Feature Alignment Network (CFAN) to gradually align discriminative features across domains through selecting effective samples from the target domain and minimizing intra-class differences. Specifically, we developed the Curriculum Transfer Strategy (CTS) and Adaptive Centroid Alignment (ACA) steps to train our model iteratively. We validated the method using three independent public ES datasets, and experimental results demonstrate that our method achieves better performance in ES classification compared with commonly used deep learning methods and existing deep domain adaptation methods.
近年来，深度学习方法在上皮-间质 (ES) 分类任务中受到了更多的关注。传统的深度学习方法假设训练数据和测试数据具有相同的分布，而这种假设在复杂的成像过程中很少得到满足。无监督领域自适应 (UDA) 将知识从有标签的源域转移到完全无标签的目标域，更适合 ES 分类任务，以避免繁琐的注释。然而，现有的用于此任务的 UDA 方法忽略了跨域的语义对齐。在本文中，我们提出了一种课程特征对齐网络 (CFAN)，通过从目标域中选择有效样本并最小化类内差异，逐步在域之间对齐判别特征。具体来说，我们开发了课程迁移策略 (CTS) 和自适应质心对齐 (ACA) 步骤来迭代训练我们的模型。我们使用三个独立的公共 ES 数据集验证了该方法，实验结果表明，与常用的深度学习方法和现有的深度域自适应方法相比，我们的方法在 ES 分类中具有更好的性能。
