Key Laboratory of Big Data-Based Precision Medicine, Ministry of Industry and Information Technology, School of Engineering Medicine, Beihang University, China; CAS Key Laboratory of Molecular Imaging, Institute of Automation, Chinese Academy of Sciences, China.
Key Laboratory of Big Data-Based Precision Medicine, Ministry of Industry and Information Technology, School of Engineering Medicine, Beihang University, China; CAS Key Laboratory of Molecular Imaging, Institute of Automation, Chinese Academy of Sciences, China.
Med Image Anal. 2024 Oct;97:103281. doi: 10.1016/j.media.2024.103281. Epub 2024 Jul 25.
Imbalanced classification is a common and difficult task in many medical image analysis applications. However, most existing approaches focus on balancing feature distribution and classifier weights between classes, while ignoring the inner-class heterogeneity and the individuality of each sample. In this paper, we proposed a sample-specific fine-grained prototype learning (SFPL) method to learn the fine-grained representation of the majority class and learn a cosine classifier specifically for each sample such that the classification model is highly tuned to the individual's characteristic. SFPL first builds multiple prototypes to represent the majority class, and then updates the prototypes through a mixture weighting strategy. Moreover, we proposed a uniform loss based on set representations to make the fine-grained prototypes distribute uniformly. To establish associations between fine-grained prototypes and cosine classifier, we propose a selective attention aggregation module to select the effective fine-grained prototypes for final classification. Extensive experiments on three different tasks demonstrate that SFPL outperforms the state-of-the-art (SOTA) methods. Importantly, as the imbalance ratio increases from 10 to 100, the improvement of SFPL over SOTA methods increases from 2.2% to 2.4%; as the training data decreases from 800 to 100, the improvement of SFPL over SOTA methods increases from 2.2% to 3.8%.
不平衡分类是许多医学图像分析应用中的一个常见且具有挑战性的任务。然而，大多数现有的方法侧重于平衡类别之间的特征分布和分类器权重，而忽略了类内异质性和每个样本的个体性。在本文中，我们提出了一种样本特定的细粒度原型学习 (SFPL) 方法，以学习多数类的细粒度表示，并为每个样本学习余弦分类器，从而使分类模型高度适应个体的特征。SFPL 首先构建多个原型来表示多数类，然后通过混合加权策略更新原型。此外，我们提出了一种基于集合表示的一致损失来使细粒度原型均匀分布。为了在细粒度原型和余弦分类器之间建立关联，我们提出了一种选择性注意聚合模块，用于选择有效的细粒度原型进行最终分类。在三个不同任务上的广泛实验表明，SFPL 优于最先进的方法 (SOTA)。重要的是，随着不平衡比从 10 增加到 100，SFPL 相对于 SOTA 方法的改进从 2.2%增加到 2.4%；随着训练数据从 800 减少到 100，SFPL 相对于 SOTA 方法的改进从 2.2%增加到 3.8%。
