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

Deep learning models are sensitive to domain shift phenomena. A model trainedon images from one domain cannot generalise well when tested on images from adifferent domain, despite capturing similar anatomical structures. It is mainlybecause the data distribution between the two domains is different. Moreover,creating annotation for every new modality is a tedious and time-consumingtask, which also suffers from high inter- and intra- observer variability.Unsupervised domain adaptation (UDA) methods intend to reduce the gap betweensource and target domains by leveraging source domain labelled data to generatelabels for the target domain. However, current state-of-the-art (SOTA) UDAmethods demonstrate degraded performance when there is insufficient data insource and target domains. In this paper, we present a novel UDA method formulti-modal cardiac image segmentation. The proposed method is based onadversarial learning and adapts network features between source and targetdomain in different spaces. The paper introduces an end-to-end framework thatintegrates: a) entropy minimisation, b) output feature space alignment and c) anovel point-cloud shape adaptation based on the latent features learned by thesegmentation model. We validated our method on two cardiac datasets by adaptingfrom the annotated source domain, bSSFP-MRI (balanced Steady-State FreeProcession-MRI), to the unannotated target domain, LGE-MRI (Late-gadoliniumenhance-MRI), for the multi-sequence dataset; and from MRI (source) to CT(target) for the cross-modality dataset. The results highlighted that byenforcing adversarial learning in different parts of the network, the proposedmethod delivered promising performance, compared to other SOTA methods.

Adapt Everywhere: Unsupervised Adaptation of Point-Clouds and Entropy Minimisation for Multi-modal Cardiac Image Segmentation