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

The detection of nuclei and cells in histology images is of great value inboth clinical practice and pathological studies. However, multiple reasons suchas morphological variations of nuclei or cells make it a challenging task whereconventional object detection methods cannot obtain satisfactory performance inmany cases. A detection task consists of two sub-tasks, classification andlocalization. Under the condition of dense object detection, classification isa key to boost the detection performance. Considering this, we proposesimilarity based region proposal networks (SRPN) for nuclei and cells detectionin histology images. In particular, a customized convolution layer termed asembedding layer is designed for network building. The embedding layer is addedinto the region proposal networks, enabling the networks to learndiscriminative features based on similarity learning. Features obtained bysimilarity learning can significantly boost the classification performancecompared to conventional methods. SRPN can be easily integrated into standardconvolutional neural networks architectures such as the Faster R-CNN andRetinaNet. We test the proposed approach on tasks of multi-organ nucleidetection and signet ring cells detection in histological images. Experimentalresults show that networks applying similarity learning achieved superiorperformance on both tasks when compared to their counterparts. In particular,the proposed SRPN achieve state-of-the-art performance on the MoNuSeg benchmarkfor nuclei segmentation and detection while compared to previous methods, andon the signet ring cell detection benchmark when compared with baselines. Thesourcecode is publicly available at:https://github.com/sigma10010/nuclei_cells_det.

SRPN: similarity-based region proposal networks for nuclei and cells detection in histology images