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

Convolutional neural networks (CNNs) often suffer from poor performance whentested on target data that differs from the training (source) datadistribution, particularly in medical imaging applications where variations inimaging protocols across different clinical sites and scanners lead todifferent imaging appearances. However, re-accessing source training data forunsupervised domain adaptation or labeling additional test data for modelfine-tuning can be difficult due to privacy issues and high labeling costs,respectively. To solve this problem, we propose a novel atlas-guided test-timeadaptation (TTA) method for robust 3D medical image segmentation, calledAdaAtlas. AdaAtlas only takes one single unlabeled test sample as input andadapts the segmentation network by minimizing an atlas-based loss.Specifically, the network is adapted so that its prediction after registrationis aligned with the learned atlas in the atlas space, which helps to reduceanatomical segmentation errors at test time. In addition, different from mostexisting TTA methods which restrict the adaptation to batch normalizationblocks in the segmentation network only, we further exploit the use of channeland spatial attention blocks for improved adaptability at test time. Extensiveexperiments on multiple datasets from different sites show that AdaAtlas withattention blocks adapted (AdaAtlas-Attention) achieves superior performanceimprovements, greatly outperforming other competitive TTA methods.

Pay Attention to the Atlas: Atlas-Guided Test-Time Adaptation Method for Robust 3D Medical Image Segmentation