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

Anisotropic multi-slice Cardiac Magnetic Resonance (CMR) Images areconventionally acquired in patient-specific short-axis (SAX) orientation. Inspecific cardiovascular diseases that affect right ventricular (RV) morphology,acquisitions in standard axial (AX) orientation are preferred by someinvestigators, due to potential superiority in RV volume measurement fortreatment planning. Unfortunately, due to the rare occurrence of thesediseases, data in this domain is scarce. Recent research in deep learning-basedmethods mainly focused on SAX CMR images and they had proven to be verysuccessful. In this work, we show that there is a considerable domain shiftbetween AX and SAX images, and therefore, direct application of existing modelsyield sub-optimal results on AX samples. We propose a novel unsupervised domainadaptation approach, which uses task-related probabilities in an attentionmechanism. Beyond that, cycle consistency is imposed on the learnedpatient-individual 3D rigid transformation to improve stability whenautomatically re-sampling the AX images to SAX orientations. The network wastrained on 122 registered 3D AX-SAX CMR volume pairs from a multi-centricpatient cohort. A mean 3D Dice of $0.86\pm{0.06}$ for the left ventricle,$0.65\pm{0.08}$ for the myocardium, and $0.77\pm{0.10}$ for the right ventriclecould be achieved. This is an improvement of $25\%$ in Dice for RV incomparison to direct application on axial slices. To conclude, our pre-trainedtask module has neither seen CMR images nor labels from the target domain, butis able to segment them after the domain gap is reduced. Code:https://github.com/Cardio-AI/3d-mri-domain-adaptation

Unsupervised Domain Adaptation From Axial to Short-Axis Multi-Slice Cardiac MR Images by Incorporating Pretrained Task Networks