Adaptive Label Correction for Robust Medical Image Segmentation with Noisy Labels

2025-03-15

AI Summary

To address performance degradation in medical image segmentation caused by noisy annotations, this paper proposes a Mean Teacher–based self-ensembling framework with adaptive label correction. Methodologically, it introduces a dynamic multi-perturbation label weighting refinement mechanism and a sample-level uncertainty–driven progressive label filtering strategy—where uncertainty is quantified via prediction variance—integrated with consistency regularization and perturbation-robust training to enable adaptive noise correction and effective utilization of noisy labels. Experiments on two public medical imaging datasets demonstrate that the proposed method achieves an average Dice score improvement of 4.2% over state-of-the-art approaches. Notably, it maintains strong robustness under high-noise conditions (label error rates ≥ 40%), outperforming existing methods in both accuracy and stability.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Stochastic 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

Deep learning has shown remarkable success in medical image analysis, but its reliance on large volumes of high-quality labeled data limits its applicability. While noisy labeled data are easier to obtain, directly incorporating them into training can degrade model performance. To address this challenge, we propose a Mean Teacher-based Adaptive Label Correction (ALC) self-ensemble framework for robust medical image segmentation with noisy labels. The framework leverages the Mean Teacher architecture to ensure consistent learning under noise perturbations. It includes an adaptive label refinement mechanism that dynamically captures and weights differences across multiple disturbance versions to enhance the quality of noisy labels. Additionally, a sample-level uncertainty-based label selection algorithm is introduced to prioritize high-confidence samples for network updates, mitigating the impact of noisy annotations. Consistency learning is integrated to align the predictions of the student and teacher networks, further enhancing model robustness. Extensive experiments on two public datasets demonstrate the effectiveness of the proposed framework, showing significant improvements in segmentation performance. By fully exploiting the strengths of the Mean Teacher structure, the ALC framework effectively processes noisy labels, adapts to challenging scenarios, and achieves competitive results compared to state-of-the-art methods.

Problem

• Address noisy labels in medical image segmentation
• Enhance label quality using adaptive refinement
• Improve model robustness with consistency learning

Innovation

• Mean Teacher-based Adaptive Label Correction framework
• Adaptive label refinement mechanism for noisy labels
• Sample-level uncertainty-based label selection algorithm