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/

Zendy Plus

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

Zendy Tools

Zendy Open

Detectingandsegmentingobjectswithinwholeslideimagesis essential incomputational pathology workflow. Self-supervised learning (SSL) is appealingto such annotation-heavy tasks. Despite the extensive benchmarks in naturalimages for dense tasks, such studies are, unfortunately, absent in currentworks for pathology. Our paper intends to narrow this gap. We first benchmarkrepresentative SSL methods for dense prediction tasks in pathology images.Then, we propose concept contrastive learning (ConCL), an SSL framework fordense pre-training. We explore how ConCL performs with concepts provided bydifferent sources and end up with proposing a simple dependency-free conceptgenerating method that does not rely on external segmentation algorithms orsaliency detection models. Extensive experiments demonstrate the superiority ofConCL over previous state-of-the-art SSL methods across different settings.Along our exploration, we distll several important and intriguing componentscontributing to the success of dense pre-training for pathology images. We hopethis work could provide useful data points and encourage the community toconduct ConCL pre-training for problems of interest. Code is available.

ConCL: Concept Contrastive Learning for Dense Prediction Pre-training in Pathology Images