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 artery/vein (A/V) classification is a critical technique fordiagnosing diabetes and cardiovascular diseases. Although deep learning basedmethods achieve impressive results in A/V classification, their performancesusually degrade severely when being directly applied to another database, dueto the domain shift, e.g., caused by the variations in imaging protocols. Inthis paper, we propose a novel vessel-mixing based consistency regularizationframework, for cross-domain learning in retinal A/V classification. Specially,to alleviate the severe bias to source domain, based on the label smooth prior,the model is regularized to give consistent predictions for unlabeledtarget-domain inputs that are under perturbation. This consistencyregularization implicitly introduces a mechanism where the model and theperturbation is opponent to each other, where the model is pushed to be robustenough to cope with the perturbation. Thus, we investigate a more difficultopponent to further inspire the robustness of model, in the scenario of retinalA/V, called vessel-mixing perturbation. Specially, it effectively disturbs thefundus images especially the vessel structures by mixing two images regionally.We conduct extensive experiments on cross-domain A/V classification using fourpublic datasets, which are collected by diverse institutions and imagingdevices. The results demonstrate that our method achieves the state-of-the-artcross-domain performance, which is also close to the upper bound obtained byfully supervised learning on target domain.

Consistent Posterior Distributions under Vessel-Mixing: A Regularization for Cross-Domain Retinal Artery/Vein Classification