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

One of the biggest challenges for applying machine learning to histopathologyis weak supervision: whole-slide images have billions of pixels yet often onlyone global label. The state of the art therefore relies on strongly-supervisedmodel training using additional local annotations from domain experts. However,in the absence of detailed annotations, most weakly-supervised approachesdepend on a frozen feature extractor pre-trained on ImageNet. We identify thisas a key weakness and propose to train an in-domain feature extractor onhistology images using MoCo v2, a recent self-supervised learning algorithm.Experimental results on Camelyon16 and TCGA show that the proposed extractorgreatly outperforms its ImageNet counterpart. In particular, our resultsimprove the weakly-supervised state of the art on Camelyon16 from 91.4% to98.7% AUC, thereby closing the gap with strongly-supervised models that reach99.3% AUC. Through these experiments, we demonstrate that feature extractorstrained via self-supervised learning can act as drop-in replacements tosignificantly improve existing machine learning techniques in histology.Lastly, we show that the learned embedding space exhibits biologicallymeaningful separation of tissue structures.

Self-Supervision Closes the Gap Between Weak and Strong Supervision in Histology