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

Artificial intelligence (AI) models have been developed for predictingclinically relevant biomarkers, including microsatellite instability (MSI), forcolorectal cancers (CRC). However, the current deep-learning networks aredata-hungry and require large training datasets, which are often lacking in themedical domain. In this study, based on the latest Hierarchical VisionTransformer using Shifted Windows (Swin-T), we developed an efficient workflowfor biomarkers in CRC (MSI, hypermutation, chromosomal instability, CpG islandmethylator phenotype, BRAF, and TP53 mutation) that only required relativelysmall datasets, but achieved the state-of-the-art (SOTA) predictiveperformance. Our Swin-T workflow not only substantially outperformed publishedmodels in an intra-study cross-validation experiment using TCGA-CRC-DX dataset(N = 462), but also showed excellent generalizability in cross-study externalvalidation and delivered a SOTA AUROC of 0.90 for MSI using the MCO dataset fortraining (N = 1065) and the same TCGA-CRC-DX for testing. Similar performance(AUROC=0.91) was achieved by Echle and colleagues using approximately 8000training samples (ResNet18) on the same testing dataset. Swin-T was extremelyefficient using small training datasets and exhibits robust predictiveperformance with only 200-500 training samples. These data indicate that Swin-Tmay be 5-10 times more efficient than the current state-of-the-art algorithmsfor MSI based on ResNet18 and ShuffleNet. Furthermore, the Swin-T models showedpromise as pre-screening tests for MSI status and BRAF mutation status, whichcould exclude and reduce the samples before the subsequent standard testing ina cascading diagnostic workflow to allow turnaround time reduction and costsaving.

Predicting microsatellite instability and key biomarkers in colorectal  cancer from H&E-stained images: Achieving SOTA predictive performance with  fewer data using Swin Transformer