Decoupled Competitive Framework for Semi-Supervised Medical Image Segmentation

2025-05-30

European Conference on Artificial Intelligence

AI Summary

Medical image segmentation suffers from scarce annotated data, while existing semi-supervised methods—such as Mean Teacher and Dual Students—are hindered by excessive teacher-student coupling or inter-student cognitive discrepancies. To address these limitations, this paper proposes a Decoupled Competition Framework (DCF). DCF introduces a novel dynamic competitive Exponential Moving Average (EMA) update mechanism to alleviate over-reliance between teacher and student models, and designs a high-fidelity cross-student feature distillation pathway to mitigate inter-student cognitive bias. The framework integrates consistency regularization, multi-view collaborative training, and bidirectional knowledge distillation across students. Evaluated on three public 2D/3D medical imaging datasets, DCF consistently outperforms state-of-the-art semi-supervised approaches—including Mean Teacher, UDA, and CPS—with Dice score improvements of 1.8–3.2 percentage points.

Technology Category

• Machine Learning: Semi-Supervised Learning
• Computer Vision: Segmentation
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Confronting the critical challenge of insufficiently annotated samples in medical domain, semi-supervised medical image segmentation (SSMIS) emerges as a promising solution. Specifically, most methodologies following the Mean Teacher (MT) or Dual Students (DS) architecture have achieved commendable results. However, to date, these approaches face a performance bottleneck due to two inherent limitations, extit{e.g.}, the over-coupling problem within MT structure owing to the employment of exponential moving average (EMA) mechanism, as well as the severe cognitive bias between two students of DS structure, both of which potentially lead to reduced efficacy, or even model collapse eventually. To mitigate these issues, a Decoupled Competitive Framework (DCF) is elaborated in this work, which utilizes a straightforward competition mechanism for the update of EMA, effectively decoupling students and teachers in a dynamical manner. In addition, the seamless exchange of invaluable and precise insights is facilitated among students, guaranteeing a better learning paradigm. The DCF introduced undergoes rigorous validation on three publicly accessible datasets, which encompass both 2D and 3D datasets. The results demonstrate the superiority of our method over previous cutting-edge competitors. Code will be available at https://github.com/JiaheChen2002/DCF.

Problem

• Over-coupling in Mean Teacher due to EMA mechanism
• Cognitive bias between Dual Students in DS structure
• Performance bottleneck in semi-supervised medical image segmentation

Innovation

• Decoupled Competitive Framework for segmentation
• Dynamic EMA update via competition mechanism
• Facilitates seamless insight exchange among students