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

We consider the problem of segmenting cell nuclei instances from Hematoxylinand Eosin (H&E) stains with weak supervision. While most recent works focus onimproving the segmentation quality, this is usually insufficient for instancesegmentation of cell instances clumped together or with a small size. In thiswork, we propose a two-step post-processing procedure, Split and Expand, thatdirectly improves the conversion of segmentation maps to instances. In theSplit step, we split clumps of cells from the segmentation map into individualcell instances with the guidance of cell-center predictions through GaussianMixture Model clustering. In the Expand step, we find missing small cells usingthe cell-center predictions (which tend to capture small cells moreconsistently as they are trained using reliable point annotations), and utilizeLayer-wise Relevance Propagation (LRP) explanation results to expand thosecell-center predictions into cell instances. Our Split and Expandpost-processing procedure is training-free and is executed at inference-timeonly. To further improve the performance of our method, a feature re-weightingloss based on LRP is proposed. We test our procedure on the MoNuSeg and TNBCdatasets and show that our proposed method provides statistically significantimprovements on object-level metrics. Our code will be made available.

Split and Expand: An inference-time improvement for Weakly Supervised Cell Instance Segmentation