EAUWSeg: Eliminating annotation uncertainty in weakly-supervised medical image segmentation

2025-01-03

AI Summary

Weakly supervised medical image segmentation suffers from performance degradation compared to fully supervised methods due to uncertainty inherent in coarse annotations. Method: This paper proposes Bounded Polygon Annotation (BPAnno), a novel annotation paradigm, and EAUWSeg, an end-to-end weakly supervised framework. It innovatively decouples a single boundary annotation into two complementary supervision signals—inner boundary (definitively foreground) and outer boundary (definitively background). An adversarial supervision mechanism is introduced to learn lesion-invariant features, while a classification-guided confidence generator enhances supervision reliability in ambiguous boundary regions. Contribution/Results: EAUWSeg achieves state-of-the-art performance across multiple medical segmentation benchmarks, surpassing existing weakly supervised approaches and even outperforming fully supervised baselines in several cases. Crucially, it reduces annotation cost by over 80%, establishing a new paradigm for high-accuracy, low-cost clinical segmentation assistance.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Semi-Supervised Learning
• Reasoning under Uncertainty: Other Foundations of Reasoning under Uncertainty

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Semantics and Knowledge: Representation, semantic annotation, enhancement, enrichments, access and/or integration of a variety of data on the Web
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Weakly-supervised medical image segmentation is gaining traction as it requires only rough annotations rather than accurate pixel-to-pixel labels, thereby reducing the workload for specialists. Although some progress has been made, there is still a considerable performance gap between the label-efficient methods and fully-supervised one, which can be attributed to the uncertainty nature of these weak labels. To address this issue, we propose a novel weak annotation method coupled with its learning framework EAUWSeg to eliminate the annotation uncertainty. Specifically, we first propose the Bounded Polygon Annotation (BPAnno) by simply labeling two polygons for a lesion. Then, the tailored learning mechanism that explicitly treat bounded polygons as two separated annotations is proposed to learn invariant feature by providing adversarial supervision signal for model training. Subsequently, a confidence-auxiliary consistency learner incorporates with a classification-guided confidence generator is designed to provide reliable supervision signal for pixels in uncertain region by leveraging the feature presentation consistency across pixels within the same category as well as class-specific information encapsulated in bounded polygons annotation. Experimental results demonstrate that EAUWSeg outperforms existing weakly-supervised segmentation methods. Furthermore, compared to fully-supervised counterparts, the proposed method not only delivers superior performance but also costs much less annotation workload. This underscores the superiority and effectiveness of our approach.

Problem

• Weakly Supervised Learning
• Medical Image Segmentation
• Annotation Uncertainty

Innovation

• Weakly Supervised Learning
• Medical Image Segmentation
• Invariant Feature Recognition