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/

Zendy Plus

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

Zendy Tools

Zendy Open

Segmentation of brain arterio-venous malformations (bAVMs) in 3D rotationalangiographies (3DRA) is still an open problem in the literature, with highrelevance for clinical practice. While deep learning models have been appliedfor segmenting the brain vasculature in these images, they have never been usedin cases with bAVMs. This is likely caused by the difficulty to obtainsufficiently annotated data to train these approaches. In this paper weintroduce a first deep learning model for blood vessel segmentation in 3DRAimages of patients with bAVMs. To this end, we densely annotated 5 3DRA volumesof bAVM cases and used these to train two alternative 3DUNet-basedarchitectures with different segmentation objectives. Our results show that thenetworks reach a comprehensive coverage of relevant structures for bAVManalysis, much better than what is obtained using standard methods. This ispromising for achieving a better topological and morphological characterisationof the bAVM structures of interest. Furthermore, the models have the ability tosegment venous structures even when missing in the ground truth labelling,which is relevant for planning interventional treatments. Ultimately, theseresults could be used as more reliable first initial guesses, alleviating thecumbersome task of creating manual labels.

A deep learning model for brain vessel segmentation in 3DRA with arteriovenous malformations