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

Automatic multiple sclerosis (MS) lesion segmentation using multi-contrastmagnetic resonance (MR) images provides improved efficiency and reproducibilitycompared to manual delineation. Current state-of-the-art automatic MS lesionsegmentation methods utilize modified U-Net-like architectures. However, in theliterature, dedicated architecture modifications were always required tomaximize their performance. In addition, the best-performing methods have notproven to be generalizable to diverse test datasets with contrast variationsand image artifacts. In this work, we developed an accurate and generalizableMS lesion segmentation model using the well-known U-Net architecture withoutfurther modification. A novel test-time self-ensembled lesion fusion strategyis proposed that not only achieved the best performance using the ISBI 2015 MSsegmentation challenge data but also demonstrated robustness across variousself-ensemble parameter choices. Moreover, equipped with instance normalizationrather than batch normalization widely used in literature, the model trained onthe ISBI challenge data generalized well on clinical test datasets fromdifferent scanners.

Towards an accurate and generalizable multiple sclerosis lesion segmentation model using self-ensembled lesion fusion