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 Plus annual

Presents the access of premium content as premium feature

Premium Content

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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

U-Net has been providing state-of-the-art performance in many medical imagesegmentation problems. Many modifications have been proposed for U-Net, such asattention U-Net, recurrent residual convolutional U-Net (R2-UNet), and U-Netwith residual blocks or blocks with dense connections. However, all thesemodifications have an encoder-decoder structure with skip connections, and thenumber of paths for information flow is limited. We propose LadderNet in thispaper, which can be viewed as a chain of multiple U-Nets. Instead of only onepair of encoder branch and decoder branch in U-Net, a LadderNet has multiplepairs of encoder-decoder branches, and has skip connections between every pairof adjacent decoder and decoder branches in each level. Inspired by the successof ResNet and R2-UNet, we use modified residual blocks where two convolutionallayers in one block share the same weights. A LadderNet has more paths forinformation flow because of skip connections and residual blocks, and can beviewed as an ensemble of Fully Convolutional Networks (FCN). The equivalence toan ensemble of FCNs improves segmentation accuracy, while the shared weightswithin each residual block reduce parameter number. Semantic segmentation isessential for retinal disease detection. We tested LadderNet on two benchmarkdatasets for blood vessel segmentation in retinal images, and achieved superiorperformance over methods in the literature. The implementation is provided\url{https://github.com/juntang-zhuang/LadderNet}

LadderNet: Multi-path networks based on U-Net for medical image segmentation