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

Annotation scarcity is a long-standing problem in medical image analysisarea. To efficiently leverage limited annotations, abundant unlabeled data areadditionally exploited in semi-supervised learning, while well-establishedcross-modality data are investigated in domain adaptation. In this paper, weaim to explore the feasibility of concurrently leveraging both unlabeled dataand cross-modality data for annotation-efficient cardiac segmentation. To thisend, we propose a cutting-edge semi-supervised domain adaptation framework,namely Dual-Teacher++. Besides directly learning from limited labeled targetdomain data (e.g., CT) via a student model adopted by previous literature, wedesign novel dual teacher models, including an inter-domain teacher model toexplore cross-modality priors from source domain (e.g., MR) and an intra-domainteacher model to investigate the knowledge beneath unlabeled target domain. Inthis way, the dual teacher models would transfer acquired inter- andintra-domain knowledge to the student model for further integration andexploitation. Moreover, to encourage reliable dual-domain knowledge transfer,we enhance the inter-domain knowledge transfer on the samples with highersimilarity to target domain after appearance alignment, and also strengthenintra-domain knowledge transfer of unlabeled target data with higher predictionconfidence. In this way, the student model can obtain reliable dual-domainknowledge and yield improved performance on target domain data. We extensivelyevaluated the feasibility of our method on the MM-WHS 2017 challenge dataset.The experiments have demonstrated the superiority of our framework over othersemi-supervised learning and domain adaptation methods. Moreover, ourperformance gains could be yielded in bidirections,i.e., adapting from MR toCT, and from CT to MR.

Dual-Teacher++: Exploiting Intra-domain and Inter-domain Knowledge with Reliable Transfer for Cardiac Segmentation