English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Automatic polyp segmentation plays a crucial role in the early diagnosis andtreatment of colorectal cancer (CRC). However, existing methods heavily rely onfully supervised training, which requires a large amount of labeled data withtime-consuming pixel-wise annotations. Moreover, accurately segmenting polypsposes challenges due to variations in shape, size, and location. To addressthese issues, we propose a novel Dual-scale Enhanced and Cross-generativeconsistency learning framework for semi-supervised polyp Segmentation (DEC-Seg)from colonoscopy images. First, we propose a Cross-level Feature Aggregation(CFA) module that integrates cross-level adjacent layers to enhance the featurerepresentation ability across different resolutions. To address scalevariation, we present a scale-enhanced consistency constraint, which ensuresconsistency in the segmentation maps generated from the same input image atdifferent scales. This constraint helps handle variations in polyp sizes andimproves the robustness of the model. Additionally, we design a scale-awareperturbation consistency scheme to enhance the robustness of the mean teachermodel. Furthermore, we propose a cross-generative consistency scheme, in whichthe original and perturbed images can be reconstructed using cross-segmentationmaps. This consistency constraint allows us to mine effective featurerepresentations and boost the segmentation performance. To produce moreaccurate segmentation maps, we propose a Dual-scale Complementary Fusion (DCF)module that integrates features from two scale-specific decoders operating atdifferent scales. Extensive experimental results on five benchmark datasetsdemonstrate the effectiveness of our DEC-Seg against other state-of-the-artsemi-supervised segmentation approaches. The implementation code will bereleased at https://github.com/taozh2017/DECSeg.

Dual-scale Enhanced and Cross-generative Consistency Learning for Semi-supervised Polyp Segmentation