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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

In computational pathology, multiple instance learning (MIL) is widely usedto circumvent the computational impasse in giga-pixel whole slide image (WSI)analysis. It usually consists of two stages: patch-level feature extraction andslide-level aggregation. Recently, pretrained models or self-supervisedlearning have been used to extract patch features, but they suffer from loweffectiveness or inefficiency due to overlooking the task-specific supervisionprovided by slide labels. Here we propose a weakly-supervised Label-EfficientWSI Screening method, dubbed LESS, for cytological WSI analysis with onlyslide-level labels, which can be effectively applied to small datasets. First,we suggest using variational positive-unlabeled (VPU) learning to uncoverhidden labels of both benign and malignant patches. We provide appropriatesupervision by using slide-level labels to improve the learning of patch-levelfeatures. Next, we take into account the sparse and random arrangement of cellsin cytological WSIs. To address this, we propose a strategy to crop patches atmultiple scales and utilize a cross-attention vision transformer (CrossViT) tocombine information from different scales for WSI classification. Thecombination of our two steps achieves task-alignment, improving effectivenessand efficiency. We validate the proposed label-efficient method on a urinecytology WSI dataset encompassing 130 samples (13,000 patches) and FNAC 2019dataset with 212 samples (21,200 patches). The experiment shows that theproposed LESS reaches 84.79%, 85.43%, 91.79% and 78.30% on a urine cytology WSIdataset, and 96.88%, 96.86%, 98.95%, 97.06% on FNAC 2019 dataset in terms ofaccuracy, AUC, sensitivity and specificity. It outperforms state-of-the-art MILmethods on pathology WSIs and realizes automatic cytological WSI cancerscreening.

LESS: Label-efficient Multi-scale Learning for Cytological Whole Slide Image Screening