English

ACTION++: Improving Semi-supervised Medical Image Segmentation with Adaptive Anatomical Contrast

Computer Vision and Pattern Recognition 2023-07-19 v3 Artificial Intelligence Machine Learning Image and Video Processing

Abstract

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 τ\tau in the contrastive loss on long-tailed medical data is not optimal, and propose to use a dynamic τ\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.

Keywords

Cite

@article{arxiv.2304.02689,
  title  = {ACTION++: Improving Semi-supervised Medical Image Segmentation with Adaptive Anatomical Contrast},
  author = {Chenyu You and Weicheng Dai and Yifei Min and Lawrence Staib and Jasjeet S. Sekhon and James S. Duncan},
  journal= {arXiv preprint arXiv:2304.02689},
  year   = {2023}
}

Comments

Accepted by International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI 2023)

R2 v1 2026-06-28T09:51:40.970Z