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 present a semi-supervised domain adaptation framework for brain vesselsegmentation from different image modalities. Existing state-of-the-art methodsfocus on a single modality, despite the wide range of available cerebrovascularimaging techniques. This can lead to significant distribution shifts thatnegatively impact the generalization across modalities. By relying on annotatedangiographies and a limited number of annotated venographies, our frameworkaccomplishes image-to-image translation and semantic segmentation, leveraging adisentangled and semantically rich latent space to represent heterogeneous dataand perform image-level adaptation from source to target domains. Moreover, wereduce the typical complexity of cycle-based architectures and minimize the useof adversarial training, which allows us to build an efficient and intuitivemodel with stable training. We evaluate our method on magnetic resonanceangiographies and venographies. While achieving state-of-the-art performance inthe source domain, our method attains a Dice score coefficient in the targetdomain that is only 8.9% lower, highlighting its promising potential for robustcerebrovascular image segmentation across different modalities.

A2V: A Semi-Supervised Domain Adaptation Framework for Brain Vessel Segmentation via Two-Phase Training Angiography-to-Venography Translation