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

Although the U-Net architecture has been extensively used for segmentation ofmedical images, we address two of its shortcomings in this work. Firstly, theaccuracy of vanilla U-Net degrades when the target regions for segmentationexhibit significant variations in shape and size. Even though the U-Net alreadypossesses some capability to analyze features at various scales, we propose toexplicitly add multi-scale feature maps in each convolutional module of theU-Net encoder to improve segmentation of histology images. Secondly, theaccuracy of a U-Net model also suffers when the annotations for supervisedlearning are noisy or incomplete. This can happen due to the inherentdifficulty for a human expert to identify and delineate all instances ofspecific pathology very precisely and accurately. We address this challenge byintroducing auxiliary confidence maps that emphasize less on the boundaries ofthe given target regions. Further, we utilize the bootstrapping properties ofthe deep network to address the missing annotation problem intelligently. Inour experiments on a private dataset of breast cancer lymph nodes, where theprimary task was to segment germinal centres and sinus histiocytosis, weobserved substantial improvement over a U-Net baseline based on the twoproposed augmentations.

Deep Multi-Scale U-Net Architecture and Label-Noise Robust Training Strategies for Histopathological Image Segmentation