DHC: Dual-debiased Heterogeneous Co-training Framework for Class-imbalanced Semi-supervised Medical Image Segmentation

TL;DR

This paper introduces a dual-debiased co-training framework for semi-supervised 3D medical image segmentation, effectively addressing class imbalance and improving segmentation accuracy with limited labeled data.

Contribution

The novel DHC framework employs dynamic loss weighting strategies and co-training of diverse models to mitigate data and learning biases in class-imbalanced semi-supervised segmentation.

Findings

Significant performance improvements over existing SSL methods.

Effective mitigation of class imbalance in semi-supervised segmentation.

Outperforms state-of-the-art methods on new benchmarks.

Abstract

The volume-wise labeling of 3D medical images is expertise-demanded and time-consuming; hence semi-supervised learning (SSL) is highly desirable for training with limited labeled data. Imbalanced class distribution is a severe problem that bottlenecks the real-world application of these methods but was not addressed much. Aiming to solve this issue, we present a novel Dual-debiased Heterogeneous Co-training (DHC) framework for semi-supervised 3D medical image segmentation. Specifically, we propose two loss weighting strategies, namely Distribution-aware Debiased Weighting (DistDW) and Difficulty-aware Debiased Weighting (DiffDW), which leverage the pseudo labels dynamically to guide the model to solve data and learning biases. The framework improves significantly by co-training these two diverse and accurate sub-models. We also introduce more representative benchmarks for class-imbalanced semi-supervised medical image segmentation, which can fully demonstrate the efficacy of the class-imbalance designs. Experiments show that our proposed framework brings significant improvements by using pseudo labels for debiasing and alleviating the class imbalance problem. More importantly, our method outperforms the state-of-the-art SSL methods, demonstrating the potential of our framework for the more challenging SSL setting. Code and models are available at: https://github.com/xmed-lab/DHC.