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

Alex Skovsbo Jørgensen; Jonas Emborg; Rasmus Røge; Lasse Riis Østergaard

doi:10.1007/978-3-030-00949-6_8

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

Alex Skovsbo Jørgensen, Jonas Emborg, Rasmus Røge, Lasse Riis Østergaard

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

5 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 language	English
Title of host publication	Computational 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
Editors	Zeike 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 pages	8
Publisher	Springer
Publication date	2018
Pages	61-68
ISBN (Print)	978-3-030-00948-9
ISBN (Electronic)	978-3-030-00949-6
DOIs	https://doi.org/10.1007/978-3-030-00949-6_8
Publication status	Published - 2018
Event	1st International Workshop on Computational Pathology, COMPAY 2018, and 5th International Workshop on Ophthalmic Medical Image Analysis, OMIA 2018 - Granada, Spain Duration: 16 Oct 2018 → 20 Oct 2018

Conference

Conference	1st International Workshop on Computational Pathology, COMPAY 2018, and 5th International Workshop on Ophthalmic Medical Image Analysis, OMIA 2018
Country/Territory	Spain
City	Granada
Period	16/10/2018 → 20/10/2018

Series	Lecture Notes in Computer Science
Volume	11039
ISSN	0302-9743

Keywords

Cancer
Classification
Colon
H&E stain
Machine learning

Access to Document

10.1007/978-3-030-00949-6_8

AUB Link

Search for the material in Aalborg University Library's search engine

Cite this

Jørgensen, A. S., Emborg, J., Røge, R., & Østergaard, L. R. (2018). Exploiting multiple color representations to improve colon cancer detection in whole slide H&E stains. In Z. Taylor, H. Bogunovic, D. Snead, M. K. Garvin, X. J. Chen, F. Ciompi, Y. Xu, L. Maier-Hein, M. Veta, E. Trucco, D. Stoyanov, N. Rajpoot, J. van der Laak, A. Martel, & S. McKenna (Eds.), Computational 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 (pp. 61-68). Springer. https://doi.org/10.1007/978-3-030-00949-6_8

Jørgensen, Alex Skovsbo ; Emborg, Jonas ; Røge, Rasmus et al. / Exploiting multiple color representations to improve colon cancer detection in whole slide H&E stains. Computational 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. editor / Zeike 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. Springer, 2018. pp. 61-68 (Lecture Notes in Computer Science, Vol. 11039).

@inproceedings{9223ce3c4c0746e88f0de0fe13c220df,

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

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.",

keywords = "Cancer, Classification, Colon, H&E stain, Machine learning",

author = "J{\o}rgensen, {Alex Skovsbo} and Jonas Emborg and Rasmus R{\o}ge and {\O}stergaard, {Lasse Riis}",

year = "2018",

doi = "10.1007/978-3-030-00949-6_8",

language = "English",

isbn = "978-3-030-00948-9",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "61--68",

editor = "Zeike Taylor and Hrvoje Bogunovic and David Snead and Garvin, {Mona K.} and Chen, {Xin Jan} and Francesco Ciompi and Yanwu Xu and Lena Maier-Hein and Mitko Veta and Emanuele Trucco and Danail Stoyanov and Nasir Rajpoot and {van der Laak}, Jeroen and Anne Martel and Stephen McKenna",

booktitle = "Computational Pathology and Ophthalmic Medical Image Analysis",

address = "Germany",

note = "1st International Workshop on Computational Pathology, COMPAY 2018, and 5th International Workshop on Ophthalmic Medical Image Analysis, OMIA 2018 ; Conference date: 16-10-2018 Through 20-10-2018",

}

Jørgensen, AS, Emborg, J, Røge, R & Østergaard, LR 2018, Exploiting multiple color representations to improve colon cancer detection in whole slide H&E stains. in Z Taylor, H Bogunovic, D Snead, MK Garvin, XJ Chen, F Ciompi, Y Xu, L Maier-Hein, M Veta, E Trucco, D Stoyanov, N Rajpoot, J van der Laak, A Martel & S McKenna (eds), Computational 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. Springer, Lecture Notes in Computer Science, vol. 11039, pp. 61-68, 1st International Workshop on Computational Pathology, COMPAY 2018, and 5th International Workshop on Ophthalmic Medical Image Analysis, OMIA 2018, Granada, Spain, 16/10/2018. https://doi.org/10.1007/978-3-030-00949-6_8

Exploiting multiple color representations to improve colon cancer detection in whole slide H&E stains. / Jørgensen, Alex Skovsbo; Emborg, Jonas; Røge, Rasmus et al.
Computational 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. ed. / Zeike 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. Springer, 2018. p. 61-68 (Lecture Notes in Computer Science, Vol. 11039).

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

TY - GEN

T1 - Exploiting multiple color representations to improve colon cancer detection in whole slide H&E stains

AU - Jørgensen, Alex Skovsbo

AU - Emborg, Jonas

AU - Røge, Rasmus

AU - Østergaard, Lasse Riis

PY - 2018

Y1 - 2018

N2 - 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.

AB - 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.

KW - Cancer

KW - Classification

KW - Colon

KW - H&E stain

KW - Machine learning

UR - http://www.scopus.com/inward/record.url?scp=85053913472&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-00949-6_8

DO - 10.1007/978-3-030-00949-6_8

M3 - Article in proceeding

SN - 978-3-030-00948-9

T3 - Lecture Notes in Computer Science

SP - 61

EP - 68

BT - Computational Pathology and Ophthalmic Medical Image Analysis

A2 - Taylor, Zeike

A2 - Bogunovic, Hrvoje

A2 - Snead, David

A2 - Garvin, Mona K.

A2 - Chen, Xin Jan

A2 - Ciompi, Francesco

A2 - Xu, Yanwu

A2 - Maier-Hein, Lena

A2 - Veta, Mitko

A2 - Trucco, Emanuele

A2 - Stoyanov, Danail

A2 - Rajpoot, Nasir

A2 - van der Laak, Jeroen

A2 - Martel, Anne

A2 - McKenna, Stephen

PB - Springer

T2 - 1st International Workshop on Computational Pathology, COMPAY 2018, and 5th International Workshop on Ophthalmic Medical Image Analysis, OMIA 2018

Y2 - 16 October 2018 through 20 October 2018

ER -

Jørgensen AS, Emborg J, Røge R , Østergaard LR. Exploiting multiple color representations to improve colon cancer detection in whole slide H&E stains. In Taylor Z, Bogunovic H, Snead D, Garvin MK, Chen XJ, Ciompi F, Xu Y, Maier-Hein L, Veta M, Trucco E, Stoyanov D, Rajpoot N, van der Laak J, Martel A, McKenna S, editors, Computational 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. Springer. 2018. p. 61-68. (Lecture Notes in Computer Science, Vol. 11039). doi: 10.1007/978-3-030-00949-6_8

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

Abstract

Conference

Keywords

Access to Document

AUB Link

Other files and links

Fingerprint

Automated cancer diagnostics in histopathological images

Cite this

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

Abstract

Conference

Keywords

Access to Document

AUB Link

Other files and links

Fingerprint

Projects

Automated cancer diagnostics in histopathological images

Cite this