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

This paper presents a new regularization method to train a fullyconvolutional network for semantic tissue segmentation in histopathologicalimages. This method relies on the benefit of unsupervised learning, in the formof image reconstruction, for network training. To this end, it puts forward anidea of defining a new embedding that allows uniting the main supervised taskof semantic segmentation and an auxiliary unsupervised task of imagereconstruction into a single one and proposes to learn this united task by asingle generative model. This embedding generates an output image bysuperimposing an input image on its segmentation map. Then, the method learnsto translate the input image to this embedded output image using a conditionalgenerative adversarial network, which is known as quite effective forimage-to-image translations. This proposal is different than the existingapproach that uses image reconstruction for the same regularization purpose.The existing approach considers segmentation and image reconstruction as twoseparate tasks in a multi-task network, defines their losses independently, andcombines them in a joint loss function. However, the definition of such afunction requires externally determining right contributions of the supervisedand unsupervised losses that yield balanced learning between the segmentationand image reconstruction tasks. The proposed approach provides an easiersolution to this problem by uniting these two tasks into a single one, whichintrinsically combines their losses. We test our approach on three datasets ofhistopathological images. Our experiments demonstrate that it leads to bettersegmentation results in these datasets, compared to its counterparts.

Image Embedded Segmentation: Uniting Supervised and Unsupervised Objectives for Segmenting Histopathological Images