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

Degeneration of articular cartilage (AC) is actively studied in kneeosteoarthritis (OA) research via magnetic resonance imaging (MRI). Segmentationof AC tissues from MRI data is an essential step in quantification of theirdamage. Deep learning (DL) based methods have shown potential in this realm andare the current state-of-the-art, however, their robustness to heterogeneity ofMRI acquisition settings remains an open problem. In this study, weinvestigated two modern regularization techniques -- mixup and adversarialunsupervised domain adaptation (UDA) -- to improve the robustness of DL-basedknee cartilage segmentation to new MRI acquisition settings. Our validationsetup included two datasets produced by different MRI scanners and usingdistinct data acquisition protocols. We assessed the robustness of automaticsegmentation by comparing mixup and UDA approaches to a strong baseline methodat different OA severity stages and, additionally, in relation to anatomicallocations. Our results showed that for moderate changes in knee MRI dataacquisition settings both approaches may provide notable improvements in therobustness, which are consistent for all stages of the disease and affect theclinically important areas of the knee joint. However, mixup may be consideredas a recommended approach, since it is more computationally efficient and doesnot require additional data from the target acquisition setup.

Improving Robustness of Deep Learning Based Knee MRI Segmentation: Mixup and Adversarial Domain Adaptation