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

Self-supervised representation learning has been extremely successful inmedical image analysis, as it requires no human annotations to providetransferable representations for downstream tasks. Recent self-supervisedlearning methods are dominated by noise-contrastive estimation (NCE, also knownas contrastive learning), which aims to learn invariant visual representationsby contrasting one homogeneous image pair with a large number of heterogeneousimage pairs in each training step. Nonetheless, NCE-based approaches stillsuffer from one major problem that is one homogeneous pair is not enough toextract robust and invariant semantic information. Inspired by the archetypicaltriplet loss, we propose GraVIS, which is specifically optimized for learningself-supervised features from dermatology images, to group homogeneousdermatology images while separating heterogeneous ones. In addition, ahardness-aware attention is introduced and incorporated to address theimportance of homogeneous image views with similar appearance instead of thosedissimilar homogeneous ones. GraVIS significantly outperforms its transferlearning and self-supervised learning counterparts in both lesion segmentationand disease classification tasks, sometimes by 5 percents under extremelylimited supervision. More importantly, when equipped with the pre-trainedweights provided by GraVIS, a single model could achieve better results thanwinners that heavily rely on ensemble strategies in the well-known ISIC 2017challenge.

GraVIS: Grouping Augmented Views from Independent Sources for Dermatology Analysis