Compete to Win: Enhancing Pseudo Labels for Barely-supervised Medical Image Segmentation

Abstract

This study investigates barely-supervised medical image segmentation where only few labeled data, i.e., single-digit cases are available. We observe the key limitation of the existing state-of-the-art semi-supervised solution cross pseudo supervision is the unsatisfactory precision of foreground classes, leading to a degenerated result under barely-supervised learning. In this paper, we propose a novel Compete-to-Win method (ComWin) to enhance the pseudo label quality. In contrast to directly using one model's predictions as pseudo labels, our key idea is that high-quality pseudo labels should be generated by comparing multiple confidence maps produced by different networks to select the most confident one (a compete-to-win strategy). To further refine pseudo labels at near-boundary areas, an enhanced version of ComWin, namely, ComWin+, is proposed by integrating a boundary-aware enhancement module. Experiments show that our method can achieve the best performance on three public medical image datasets for cardiac structure segmentation, pancreas segmentation and colon tumor segmentation, respectively. The source code is now available at https://github.com/Huiimin5/comwin.

arXiv commentAccepted by TMI (in IEEE Transactions on Neural Networks and Learning Systems). Code available at https://github.com/Huiimin5/comwin

References in corpus (9)

• Swin-Unet: Unet-like Pure Transformer for Medical Image Segmentation
• Classification with Noisy Labels by Importance Reweighting
• A large annotated medical image dataset for the development and evaluation of segmentation algorithms
• Regularization With Stochastic Transformations and Perturbations for Deep Semi-Supervised Learning
• Shape-aware Semi-supervised 3D Semantic Segmentation for Medical Images
• FreeMatch: Self-adaptive Thresholding for Semi-supervised Learning
• Semi-Supervised Medical Image Segmentation via Cross Teaching between CNN and Transformer
• Theoretical Comparisons of Positive-Unlabeled Learning against Positive-Negative Learning
• Translation Consistent Semi-supervised Segmentation for 3D Medical Images

Cited by in corpus (3)

• Semi-Supervised Semantic Segmentation Based on Pseudo-Labels: A Survey
• Deep Learning for Pancreas Segmentation: a Systematic Review
• Adaptive Mix for Semi-Supervised Medical Image Segmentation