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

Whole slide image (WSI) has been widely used to assist automated diagnosisunder the deep learning fields. However, most previous works only discuss theSINGLE task setting which is not aligned with real clinical setting, wherepathologists often conduct multiple diagnosis tasks simultaneously. Also, it iscommonly recognized that the multi-task learning paradigm can improve learningefficiency by exploiting commonalities and differences across multiple tasks.To this end, we present a novel multi-task framework (i.e., MulGT) for WSIanalysis by the specially designed Graph-Transformer equipped with Task-awareKnowledge Injection and Domain Knowledge-driven Graph Pooling modules.Basically, with the Graph Neural Network and Transformer as the buildingcommons, our framework is able to learn task-agnostic low-level localinformation as well as task-specific high-level global representation.Considering that different tasks in WSI analysis depend on different featuresand properties, we also design a novel Task-aware Knowledge Injection module totransfer the task-shared graph embedding into task-specific feature spaces tolearn more accurate representation for different tasks. Further, we elaboratelydesign a novel Domain Knowledge-driven Graph Pooling module for each task toimprove both the accuracy and robustness of different tasks by leveragingdifferent diagnosis patterns of multiple tasks. We evaluated our method on twopublic WSI datasets from TCGA projects, i.e., esophageal carcinoma and kidneycarcinoma. Experimental results show that our method outperforms single-taskcounterparts and the state-of-theart methods on both tumor typing and stagingtasks.

MulGT: Multi-task Graph-Transformer with Task-aware Knowledge Injection and Domain Knowledge-driven Pooling for Whole Slide Image Analysis