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

We introduce a one-shot segmentation method to alleviate the burden of manualannotation for medical images. The main idea is to treat one-shot segmentationas a classical atlas-based segmentation problem, where voxel-wisecorrespondence from the atlas to the unlabelled data is learned. Subsequently,segmentation label of the atlas can be transferred to the unlabelled data withthe learned correspondence. However, since ground truth correspondence betweenimages is usually unavailable, the learning system must be well-supervised toavoid mode collapse and convergence failure. To overcome this difficulty, weresort to the forward-backward consistency, which is widely used incorrespondence problems, and additionally learn the backward correspondencesfrom the warped atlases back to the original atlas. This cycle-correspondencelearning design enables a variety of extra, cycle-consistency-based supervisionsignals to make the training process stable, while also boost the performance.We demonstrate the superiority of our method over both deep learning-basedone-shot segmentation methods and a classical multi-atlas segmentation methodvia thorough experiments.

LT-Net: Label Transfer by Learning Reversible Voxel-wise Correspondence for One-shot Medical Image Segmentation