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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Congenital Heart Disease (CHD) is a group of cardiac malformations presentalready during fetal life, representing the prevailing category of birthdefects globally. Our aim in this study is to aid 3D fetal vessel topologyvisualisation in aortic arch anomalies, a group which encompasses a range ofconditions with significant anatomical heterogeneity. We present a multi-taskframework for automated multi-class fetal vessel segmentation from 3D blackblood T2w MRI and anomaly classification. Our training data consists of binarymanual segmentation masks of the cardiac vessels' region in individual subjectsand fully-labelled anomaly-specific population atlases. Our framework combinesdeep learning label propagation using VoxelMorph with 3D Attention U-Netsegmentation and DenseNet121 anomaly classification. We target 11 cardiacvessels and three distinct aortic arch anomalies, including double aortic arch,right aortic arch, and suspected coarctation of the aorta. We incorporate ananomaly classifier into our segmentation pipeline, delivering a multi-taskframework with the primary motivation of correcting topological inaccuracies ofthe segmentation. The hypothesis is that the multi-task approach will encouragethe segmenter network to learn anomaly-specific features. As a secondarymotivation, an automated diagnosis tool may have the potential to enhancediagnostic confidence in a decision support setting. Our results showcase thatour proposed training strategy significantly outperforms label propagation anda network trained exclusively on propagated labels. Our classifier outperformsa classifier trained exclusively on T2w volume images, with an average balancedaccuracy of 0.99 (0.01) after joint training. Adding a classifier improves theanatomical and topological accuracy of all correctly classified double aorticarch subjects.

Multi-task learning for joint weakly-supervised segmentation and aortic arch anomaly classification in fetal cardiac MRI