Medical data often exhibits long-tail distributions with heavy class imbalance, which naturally leads to difficulty in classifying the minority classes (i.e., boundary regions or rare objects). Recent work has significantly improved semi-supervised medical image segmentation in long-tailed scenarios by equipping them with unsupervised contrastive criteria. However, it remains unclear how well they will perform in the labeled portion of data where class distribution is also highly imbalanced. In this work, we present ACTION++, an improved contrastive learning framework with adaptive anatomical contrast for semi-supervised medical segmentation. Specifically, we propose an adaptive supervised contrastive loss, where we first compute the optimal locations of class centers uniformly distributed on the embedding space (i.e., off-line), and then perform online contrastive matching training by encouraging different class features to adaptively match these distinct and uniformly distributed class centers. Moreover, we argue that blindly adopting a constant temperature $\tau$ in the contrastive loss on long-tailed medical data is not optimal, and propose to use a dynamic $\tau$ via a simple cosine schedule to yield better separation between majority and minority classes. Empirically, we evaluate ACTION++ on ACDC and LA benchmarks and show that it achieves state-of-the-art across two semi-supervised settings. Theoretically, we analyze the performance of adaptive anatomical contrast and confirm its superiority in label efficiency.

翻译：医学数据常呈现长尾分布且类别严重不平衡，这自然导致少数类（即边界区域或罕见目标）的分类困难。近期研究通过引入无监督对比准则显著提升了长尾场景下半监督医学图像分割的性能。然而，在类分布同样高度不平衡的标注数据部分，这些方法的有效性仍不明确。本文提出ACTION++——一种改进的具有自适应解剖对比的对比学习框架，用于半监督医学分割。具体而言，我们提出自适应监督对比损失：首先计算嵌入空间中均匀分布的类中心的最优位置（即离线计算），随后通过鼓励不同类别的特征自适应匹配这些离散且均匀分布的类中心进行在线对比匹配训练。此外，我们论证在长尾医学数据中盲目采用对比损失的恒定温度参数τ并非最优，并提出通过简单余弦调度使用动态τ以更好地区分多数类与少数类。实验方面，我们在ACDC和LA基准数据集上评估ACTION++，证明其在两种半监督设置下均达到最优性能。理论上，我们分析了自适应解剖对比的性能，并确认其在标签效率方面的优越性。