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

Accurately and quickly binuclear cell (BC) detection plays a significant rolein predicting the risk of leukemia and other malignant tumors. However, manualmicroscopy counting is time-consuming and lacks objectivity. Moreover, with thelimitation of staining quality and diversity of morphology features in BCmicroscopy whole slide images (WSIs), traditional image processing approachesare helpless. To overcome this challenge, we propose a two-stage detectionmethod inspired by the structure prior of BC based on deep learning, whichcascades to implement BCs coarse detection at the WSI-level and fine-grainedclassification in patch-level. The coarse detection network is a multi-taskdetection framework based on circular bounding boxes for cells detection, andcentral key points for nucleus detection. The circle representation reduces thedegrees of freedom, mitigates the effect of surrounding impurities compared tousual rectangular boxes and can be rotation invariant in WSI. Detecting keypoints in the nucleus can assist network perception and be used forunsupervised color layer segmentation in later fine-grained classification. Thefine classification network consists of a background region suppression modulebased on color layer mask supervision and a key region selection module basedon a transformer due to its global modeling capability. Additionally, anunsupervised and unpaired cytoplasm generator network is firstly proposed toexpand the long-tailed distribution dataset. Finally, experiments are performedon BC multicenter datasets. The proposed BC fine detection method outperformsother benchmarks in almost all the evaluation criteria, providing clarificationand support for tasks such as cancer screenings.

From WSI-level to Patch-level: Structure Prior Guided Binuclear Cell Fine-grained Detection