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

Retinal vessel segmentation plays a key role in computer-aided screening,diagnosis, and treatment of various cardiovascular and ophthalmic diseases.Recently, deep learning-based retinal vessel segmentation algorithms haveachieved remarkable performance. However, due to the domain shift problem, theperformance of these algorithms often degrades when they are applied to newdata that is different from the training data. Manually labeling new data foreach test domain is often a time-consuming and laborious task. In this work, weexplore unsupervised domain adaptation in retinal vessel segmentation by usingentropy-based adversarial learning and transfer normalization layer to train asegmentation network, which generalizes well across domains and requires noannotation of the target domain. Specifically, first, an entropy-basedadversarial learning strategy is developed to reduce the distributiondiscrepancy between the source and target domains while also achieving theobjective of entropy minimization on the target domain. In addition, a newtransfer normalization layer is proposed to further boost the transferabilityof the deep network. It normalizes the features of each domain separately tocompensate for the domain distribution gap. Besides, it also adaptively selectsthose feature channels that are more transferable between domains, thus furtherenhancing the generalization performance of the network. We conducted extensiveexperiments on three regular fundus image datasets and an ultra-widefieldfundus image dataset, and the results show that our approach yields significantperformance gains compared to other state-of-the-art methods.

Unsupervised Domain Adaptation for Retinal Vessel Segmentation with Adversarial Learning and Transfer Normalization