Exploiting multiple color representations to improve colon cancer detection in whole slide H&E stains

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

3 Citations (Scopus)

Abstract

Currently, colon cancer diagnosis is based on manual assessment of tissue samples stained with hematoxylin and eosin (H&E). This is a high volume, time consuming, and subjective task which could be aided by automatic cancer detection. We propose an algorithm for automatic cancer detection within WSI H&E stains using a multi class colon tissue classifier based on features extracted from 5 different color representations. Approx. 32000 tissue patches were extracted for the classifier from manual annotations of 9 representative colon tissue types from 74 WSI H&E stains. Colon tissue classifiers based on gray level or color features were trained using leave-one-out forward selection. The best colon tissue classifier was based on color texture features obtaining an average tissue precision-recall (PR) area under the curve (AUC) of 0.886 and a cancer PR-AUC of 0.950 on 20 validation WSI H&E stains.

Original languageEnglish
Title of host publicationComputational Pathology and Ophthalmic Medical Image Analysis : First International Workshop, COMPAY 2018, and 5th International Workshop, OMIA 2018, Held in Conjunction with MICCAI 2018, Granada, Spain, 16-20 September 2018, Proceedings
EditorsZeike Taylor, Hrvoje Bogunovic, David Snead, Mona K. Garvin, Xin Jan Chen, Francesco Ciompi, Yanwu Xu, Lena Maier-Hein, Mitko Veta, Emanuele Trucco, Danail Stoyanov, Nasir Rajpoot, Jeroen van der Laak, Anne Martel, Stephen McKenna
Number of pages8
PublisherSpringer
Publication date2018
Pages61-68
ISBN (Print)978-3-030-00948-9
ISBN (Electronic)978-3-030-00949-6
DOIs
Publication statusPublished - 2018
Event1st International Workshop on Computational Pathology, COMPAY 2018, and 5th International Workshop on Ophthalmic Medical Image Analysis, OMIA 2018 - Granada, Spain
Duration: 16 Oct 201820 Oct 2018

Conference

Conference1st International Workshop on Computational Pathology, COMPAY 2018, and 5th International Workshop on Ophthalmic Medical Image Analysis, OMIA 2018
Country/TerritorySpain
CityGranada
Period16/10/201820/10/2018
SeriesLecture Notes in Computer Science
Volume11039
ISSN0302-9743

Keywords

  • Cancer
  • Classification
  • Colon
  • H&E stain
  • Machine learning

Fingerprint

Dive into the research topics of 'Exploiting multiple color representations to improve colon cancer detection in whole slide H&E stains'. Together they form a unique fingerprint.

Cite this