Mind the Context: Attention-Guided Weak-to-Strong Consistency for Enhanced Semi-Supervised Medical Image Segmentation

2024-10-16

arXiv.org

AI Summary

Medical image segmentation is hindered by the scarcity of high-quality annotated data. To address the lack of domain-specific perturbation strategies for semi-supervised learning in medical imaging, we propose AIGCMatch—a weak-strong consistency framework guided by attention mechanisms. Our method introduces a channel-spatial joint attention mechanism at both the image and feature levels, enabling synergistic design of weak augmentations (e.g., flipping) and strong augmentations (e.g., CutMix, RandAugment) to enforce structural preservation and semantic enhancement under consistency regularization. Additionally, multi-scale consistency constraints are incorporated. On the ACDC dataset with only seven labeled cases, AIGCMatch achieves 90.4% Dice score, surpassing state-of-the-art methods. It also significantly outperforms baselines on ISIC-2017, demonstrating strong cross-modal generalizability and clinical applicability potential.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Semi-Supervised Learning
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories

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: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Medical image segmentation is a pivotal step in diagnostic and therapeutic processes, relying on high-quality annotated data that is often challenging and costly to obtain. Semi-supervised learning offers a promising approach to enhance model performance by leveraging unlabeled data. Although weak-to-strong consistency is a prevalent method in semi-supervised image segmentation, there is a scarcity of research on perturbation strategies specifically tailored for semi-supervised medical image segmentation tasks. To address this challenge, this paper introduces a simple yet efficient semi-supervised learning framework named Attention-Guided weak-to-strong Consistency Match (AIGCMatch). The AIGCMatch framework incorporates attention-guided perturbation strategies at both the image and feature levels to achieve weak-to-strong consistency regularization. This method not only preserves the structural information of medical images but also enhances the model's ability to process complex semantic information. Extensive experiments conducted on the ACDC and ISIC-2017 datasets have validated the effectiveness of AIGCMatch. Our method achieved a 90.4% Dice score in the 7-case scenario on the ACDC dataset, surpassing the state-of-the-art methods and demonstrating its potential and efficacy in clinical settings. Additionally, on the ISIC-2017 dataset, we significantly outperformed our baseline, indicating the robustness and generalizability of AIGCMatch across different medical image segmentation tasks.

Problem

• Improving semi-supervised medical image segmentation with limited annotated data
• Developing tailored perturbation strategies for medical image segmentation
• Enhancing model performance via attention-guided weak-to-strong consistency

Innovation

• Attention-guided weak-to-strong consistency framework
• Image and feature level perturbation strategies
• Preserves medical image structural information