Imbalanced Medical Image Segmentation with Pixel-dependent Noisy Labels

2025-01-12

IEEE Transactions on Medical Imaging

AI Summary

Medical image segmentation is commonly hindered by pixel-wise noisy labels and severe class imbalance, degrading model robustness and fairness. To address these challenges, we propose Collaborative Learning with Curriculum Selection (CLCS), the first framework to explicitly model pixel-level label noise dependency. CLCS features a curriculum-driven dynamic thresholding mechanism that adaptively selects high-quality samples, and introduces a noise-balanced loss to effectively leverage ambiguous noisy samples for improving minority-class learning. The method employs a dual-branch collaborative network, discrepancy-based losses, probabilistic voting, and the proposed loss function. Evaluated on multiple medical segmentation benchmarks, CLCS achieves substantial gains in minority-class Dice score (average +3.2%) while mitigating performance degradation induced by label noise—demonstrating strong robustness to realistic annotation noise and superior generalization capability.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling

Abstract

Accurate medical image segmentation is often hindered by noisy labels in training data, due to the challenges of annotating medical images. Prior research works addressing noisy labels tend to make class-dependent assumptions, overlooking the pixel-dependent nature of most noisy labels. Furthermore, existing methods typically apply fixed thresholds to filter out noisy labels, risking the removal of minority classes and consequently degrading segmentation performance. To bridge these gaps, our proposed framework, Collaborative Learning with Curriculum Selection (CLCS), addresses pixel-dependent noisy labels with class imbalance. CLCS advances the existing works by i) treating noisy labels as pixel-dependent and addressing them through a collaborative learning framework, and ii) employing a curriculum dynamic thresholding approach adapting to model learning progress to select clean data samples to mitigate the class imbalance issue, and iii) applying a noise balance loss to noisy data samples to improve data utilization instead of discarding them outright. Specifically, our CLCS contains two modules: Curriculum Noisy Label Sample Selection (CNS) and Noise Balance Loss (NBL). In the CNS module, we designed a two-branch network with discrepancy loss for collaborative learning so that different feature representations of the same instance could be extracted from distinct views and used to vote the class probabilities of pixels. Besides, a curriculum dynamic threshold is adopted to select clean-label samples through probability voting. In the NBL module, instead of directly dropping the suspiciously noisy labels, we further adopt a robust loss to leverage such instances to boost the performance.

Problem

• Medical Image Segmentation
• Label Noise
• Class Imbalance

Innovation

• Collaborative Learning and Curriculum Selection (CLCS)
• Dynamic Threshold Curriculum Noisy Sample Selection (CNS)
• Noise Balancing Loss (NBL)