Automated cancer stem cell recognition in H&E stained tissue using convolutional neural networks and color deconvolution

Wolfgang Aichinger; Sebastian Krappe; A. Enis Cetin; Rengul Cetin-Atalay; Aysegül Üner; Michaela Benz; Thomas Wittenberg; Marc Stamminger; Christian Münzenmayer

doi:10.1117/12.2254036

Automated cancer stem cell recognition in H&E stained tissue using convolutional neural networks and color deconvolution

Wolfgang Aichinger
, Sebastian Krappe
, A. Enis Cetin
, Rengul Cetin-Atalay
, Aysegül Üner
, Michaela Benz
, Thomas Wittenberg
, Marc Stamminger
, Christian Münzenmayer

Division of Medical Pathology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Citations (Scopus)

Abstract

The analysis and interpretation of histopathological samples and images is an important discipline in the diagnosis of various diseases, especially cancer. An important factor in prognosis and treatment with the aim of a precision medicine is the determination of so-called cancer stem cells (CSC) which are known for their resistance to chemotherapeutic treatment and involvement in tumor recurrence. Using immunohistochemistry with CSC markers like CD13, CD133 and others is one way to identify CSC. In our work we aim at identifying CSC presence on ubiquitous Hematoxilyn and Eosin (HE) staining as an inexpensive tool for routine histopathology based on their distinct morphological features. We present initial results of a new method based on color deconvolution (CD) and convolutional neural networks (CNN). This method performs favorably (accuracy 0.936) in comparison with a state-of-the-art method based on 1DSIFT and eigen-analysis feature sets evaluated on the same image database. We also show that accuracy of the CNN is improved by the CD pre-processing.

Original language	English
Title of host publication	Medical Imaging 2017
Subtitle of host publication	Digital Pathology
Editors	Metin N. Gurcan, John E. Tomaszewski
Publisher	SPIE
ISBN (Electronic)	9781510607255
DOIs	https://doi.org/10.1117/12.2254036
Publication status	Published - 2017
Event	Medical Imaging 2017: Digital Pathology - Orlando, United States Duration: 12 Feb 2017 → 13 Feb 2017

Publication series

Name	Progress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume	10140
ISSN (Print)	1605-7422

Conference

Conference	Medical Imaging 2017: Digital Pathology
Country/Territory	United States
City	Orlando
Period	12/02/17 → 13/02/17

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Keywords

Color deconvolution
Convolutional neural network
Deep learning
Digital pathology
Histopathology
Texture analysis

Access to Document

10.1117/12.2254036

Cite this

Aichinger, W., Krappe, S., Cetin, A. E., Cetin-Atalay, R., Üner, A., Benz, M., Wittenberg, T., Stamminger, M., & Münzenmayer, C. (2017). Automated cancer stem cell recognition in H&E stained tissue using convolutional neural networks and color deconvolution. In M. N. Gurcan, & J. E. Tomaszewski (Eds.), Medical Imaging 2017: Digital Pathology Article 101400N (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 10140). SPIE. https://doi.org/10.1117/12.2254036

@inproceedings{50da9ad41f3445cb8b07a3b596c8cad0,

title = "Automated cancer stem cell recognition in H\&E stained tissue using convolutional neural networks and color deconvolution",

abstract = "The analysis and interpretation of histopathological samples and images is an important discipline in the diagnosis of various diseases, especially cancer. An important factor in prognosis and treatment with the aim of a precision medicine is the determination of so-called cancer stem cells (CSC) which are known for their resistance to chemotherapeutic treatment and involvement in tumor recurrence. Using immunohistochemistry with CSC markers like CD13, CD133 and others is one way to identify CSC. In our work we aim at identifying CSC presence on ubiquitous Hematoxilyn and Eosin (HE) staining as an inexpensive tool for routine histopathology based on their distinct morphological features. We present initial results of a new method based on color deconvolution (CD) and convolutional neural networks (CNN). This method performs favorably (accuracy 0.936) in comparison with a state-of-the-art method based on 1DSIFT and eigen-analysis feature sets evaluated on the same image database. We also show that accuracy of the CNN is improved by the CD pre-processing.",

keywords = "Color deconvolution, Convolutional neural network, Deep learning, Digital pathology, Histopathology, Texture analysis",

author = "Wolfgang Aichinger and Sebastian Krappe and Cetin, \{A. Enis\} and Rengul Cetin-Atalay and Ayseg{\"u}l {\"U}ner and Michaela Benz and Thomas Wittenberg and Marc Stamminger and Christian M{\"u}nzenmayer",

note = "Publisher Copyright: {\textcopyright} 2017 SPIE.; Medical Imaging 2017: Digital Pathology ; Conference date: 12-02-2017 Through 13-02-2017",

year = "2017",

doi = "10.1117/12.2254036",

language = "English",

series = "Progress in Biomedical Optics and Imaging - Proceedings of SPIE",

publisher = "SPIE",

editor = "Gurcan, \{Metin N.\} and Tomaszewski, \{John E.\}",

booktitle = "Medical Imaging 2017",

address = "United States",

}

Aichinger, W, Krappe, S, Cetin, AE, Cetin-Atalay, R, Üner, A, Benz, M, Wittenberg, T, Stamminger, M & Münzenmayer, C 2017, Automated cancer stem cell recognition in H&E stained tissue using convolutional neural networks and color deconvolution. in MN Gurcan & JE Tomaszewski (eds), Medical Imaging 2017: Digital Pathology., 101400N, Progress in Biomedical Optics and Imaging - Proceedings of SPIE, vol. 10140, SPIE, Medical Imaging 2017: Digital Pathology, Orlando, United States, 12/02/17. https://doi.org/10.1117/12.2254036

Automated cancer stem cell recognition in H&E stained tissue using convolutional neural networks and color deconvolution. / Aichinger, Wolfgang; Krappe, Sebastian; Cetin, A. Enis et al.
Medical Imaging 2017: Digital Pathology. ed. / Metin N. Gurcan; John E. Tomaszewski. SPIE, 2017. 101400N (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 10140).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Automated cancer stem cell recognition in H&E stained tissue using convolutional neural networks and color deconvolution

AU - Aichinger, Wolfgang

AU - Krappe, Sebastian

AU - Cetin, A. Enis

AU - Cetin-Atalay, Rengul

AU - Üner, Aysegül

AU - Benz, Michaela

AU - Wittenberg, Thomas

AU - Stamminger, Marc

AU - Münzenmayer, Christian

PY - 2017

Y1 - 2017

N2 - The analysis and interpretation of histopathological samples and images is an important discipline in the diagnosis of various diseases, especially cancer. An important factor in prognosis and treatment with the aim of a precision medicine is the determination of so-called cancer stem cells (CSC) which are known for their resistance to chemotherapeutic treatment and involvement in tumor recurrence. Using immunohistochemistry with CSC markers like CD13, CD133 and others is one way to identify CSC. In our work we aim at identifying CSC presence on ubiquitous Hematoxilyn and Eosin (HE) staining as an inexpensive tool for routine histopathology based on their distinct morphological features. We present initial results of a new method based on color deconvolution (CD) and convolutional neural networks (CNN). This method performs favorably (accuracy 0.936) in comparison with a state-of-the-art method based on 1DSIFT and eigen-analysis feature sets evaluated on the same image database. We also show that accuracy of the CNN is improved by the CD pre-processing.

AB - The analysis and interpretation of histopathological samples and images is an important discipline in the diagnosis of various diseases, especially cancer. An important factor in prognosis and treatment with the aim of a precision medicine is the determination of so-called cancer stem cells (CSC) which are known for their resistance to chemotherapeutic treatment and involvement in tumor recurrence. Using immunohistochemistry with CSC markers like CD13, CD133 and others is one way to identify CSC. In our work we aim at identifying CSC presence on ubiquitous Hematoxilyn and Eosin (HE) staining as an inexpensive tool for routine histopathology based on their distinct morphological features. We present initial results of a new method based on color deconvolution (CD) and convolutional neural networks (CNN). This method performs favorably (accuracy 0.936) in comparison with a state-of-the-art method based on 1DSIFT and eigen-analysis feature sets evaluated on the same image database. We also show that accuracy of the CNN is improved by the CD pre-processing.

KW - Color deconvolution

KW - Convolutional neural network

KW - Deep learning

KW - Digital pathology

KW - Histopathology

KW - Texture analysis

UR - https://www.scopus.com/pages/publications/85020315476

U2 - 10.1117/12.2254036

DO - 10.1117/12.2254036

M3 - Conference contribution

AN - SCOPUS:85020315476

T3 - Progress in Biomedical Optics and Imaging - Proceedings of SPIE

BT - Medical Imaging 2017

A2 - Gurcan, Metin N.

A2 - Tomaszewski, John E.

PB - SPIE

T2 - Medical Imaging 2017: Digital Pathology

Y2 - 12 February 2017 through 13 February 2017

ER -

Aichinger W, Krappe S, Cetin AE, Cetin-Atalay R, Üner A, Benz M et al. Automated cancer stem cell recognition in H&E stained tissue using convolutional neural networks and color deconvolution. In Gurcan MN, Tomaszewski JE, editors, Medical Imaging 2017: Digital Pathology. SPIE. 2017. 101400N. (Progress in Biomedical Optics and Imaging - Proceedings of SPIE). doi: 10.1117/12.2254036

Automated cancer stem cell recognition in H&E stained tissue using convolutional neural networks and color deconvolution

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this