Colorectal tumor identification by transferring knowledge from pan-cytokeratin to H&E

Thomas E. Tavolara; Muhammad Khalid Khan Niazi; Wei Chen; Wendy Frankel; Metin N. Gurcan

doi:10.1117/12.2513014

Colorectal tumor identification by transferring knowledge from pan-cytokeratin to H&E

Thomas E. Tavolara, Muhammad Khalid Khan Niazi, Wei Chen, Wendy Frankel, Metin N. Gurcan

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

4 Scopus citations

Abstract

Tumor budding is a recently recognized, independent prognostic factor in colorectal cancer, but lacks a standardized assessment methodology. Although staining with pan-cytokeratin has been shown to mitigate the issue of lack of reproducible, intra-observer agreement, usage of this antibody remains expensive and is limited in clinical practice. We propose an automated image analysis framework to take advantage of the visual superiority of pan-cytokeratin and the routine use of H&E to detect and quantify tumor budding. Our framework has demonstrated promising ability to identify tumor regions of colorectal slides-92.0% accuracy, 94.5% sensitivity, and 85% specificity-across four independent datasets.

Original language	English
Title of host publication	Medical Imaging 2019
Subtitle of host publication	Digital Pathology
Editors	John E. Tomaszewski, Aaron D. Ward
Publisher	SPIE
ISBN (Electronic)	9781510625594
DOIs	https://doi.org/10.1117/12.2513014
State	Published - 2019
Event	Medical Imaging 2019: Digital Pathology - San Diego, United States Duration: Feb 20 2019 → Feb 21 2019

Publication series

Name	Progress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume	10956
ISSN (Print)	1605-7422

Conference

Conference	Medical Imaging 2019: Digital Pathology
Country/Territory	United States
City	San Diego
Period	02/20/19 → 02/21/19

Keywords

Classification
Colorectal cancer
Deep learning
Pan-cytokeratin
Tumor budding
cGAN

Access to Document

10.1117/12.2513014

Cite this

Tavolara, T. E., Niazi, M. K. K., Chen, W., Frankel, W., & Gurcan, M. N. (2019). Colorectal tumor identification by transferring knowledge from pan-cytokeratin to H&E. In J. E. Tomaszewski, & A. D. Ward (Eds.), Medical Imaging 2019: Digital Pathology Article 1095614 (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 10956). SPIE. https://doi.org/10.1117/12.2513014

@inproceedings{075846cea3844a35a306627f20dc594e,

title = "Colorectal tumor identification by transferring knowledge from pan-cytokeratin to H&E",

abstract = "Tumor budding is a recently recognized, independent prognostic factor in colorectal cancer, but lacks a standardized assessment methodology. Although staining with pan-cytokeratin has been shown to mitigate the issue of lack of reproducible, intra-observer agreement, usage of this antibody remains expensive and is limited in clinical practice. We propose an automated image analysis framework to take advantage of the visual superiority of pan-cytokeratin and the routine use of H&E to detect and quantify tumor budding. Our framework has demonstrated promising ability to identify tumor regions of colorectal slides-92.0% accuracy, 94.5% sensitivity, and 85% specificity-across four independent datasets.",

keywords = "Classification, Colorectal cancer, Deep learning, Pan-cytokeratin, Tumor budding, cGAN",

author = "Tavolara, {Thomas E.} and Niazi, {Muhammad Khalid Khan} and Wei Chen and Wendy Frankel and Gurcan, {Metin N.}",

note = "Funding Information: The project described was supported in part by Awards Number R01CA134451 (PIs: Gurcan, Lozanski), U24CA199374 (PIs: Gurcan, Madabushi, Martel), and U01 CA198945 (PI: Bilgin) from the National Cancer Institute, and OSU CCC Intramural Research (Pelotonia) Award. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Cancer Institute, or the National Institutes of Health. We would also like to thank Ms. Debbie Knight from the Ohio State University for helping us with data curation. Publisher Copyright: {\textcopyright} 2019 SPIE.; Medical Imaging 2019: Digital Pathology ; Conference date: 20-02-2019 Through 21-02-2019",

year = "2019",

doi = "10.1117/12.2513014",

language = "English",

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

publisher = "SPIE",

editor = "Tomaszewski, {John E.} and Ward, {Aaron D.}",

booktitle = "Medical Imaging 2019",

}

Tavolara, TE, Niazi, MKK, Chen, W , Frankel, W & Gurcan, MN 2019, Colorectal tumor identification by transferring knowledge from pan-cytokeratin to H&E. in JE Tomaszewski & AD Ward (eds), Medical Imaging 2019: Digital Pathology., 1095614, Progress in Biomedical Optics and Imaging - Proceedings of SPIE, vol. 10956, SPIE, Medical Imaging 2019: Digital Pathology, San Diego, United States, 02/20/19. https://doi.org/10.1117/12.2513014

Colorectal tumor identification by transferring knowledge from pan-cytokeratin to H&E. / Tavolara, Thomas E.; Niazi, Muhammad Khalid Khan; Chen, Wei et al.
Medical Imaging 2019: Digital Pathology. ed. / John E. Tomaszewski; Aaron D. Ward. SPIE, 2019. 1095614 (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 10956).

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

TY - GEN

T1 - Colorectal tumor identification by transferring knowledge from pan-cytokeratin to H&E

AU - Tavolara, Thomas E.

AU - Niazi, Muhammad Khalid Khan

AU - Chen, Wei

AU - Frankel, Wendy

AU - Gurcan, Metin N.

N1 - Funding Information: The project described was supported in part by Awards Number R01CA134451 (PIs: Gurcan, Lozanski), U24CA199374 (PIs: Gurcan, Madabushi, Martel), and U01 CA198945 (PI: Bilgin) from the National Cancer Institute, and OSU CCC Intramural Research (Pelotonia) Award. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Cancer Institute, or the National Institutes of Health. We would also like to thank Ms. Debbie Knight from the Ohio State University for helping us with data curation. Publisher Copyright: © 2019 SPIE.

PY - 2019

Y1 - 2019

N2 - Tumor budding is a recently recognized, independent prognostic factor in colorectal cancer, but lacks a standardized assessment methodology. Although staining with pan-cytokeratin has been shown to mitigate the issue of lack of reproducible, intra-observer agreement, usage of this antibody remains expensive and is limited in clinical practice. We propose an automated image analysis framework to take advantage of the visual superiority of pan-cytokeratin and the routine use of H&E to detect and quantify tumor budding. Our framework has demonstrated promising ability to identify tumor regions of colorectal slides-92.0% accuracy, 94.5% sensitivity, and 85% specificity-across four independent datasets.

AB - Tumor budding is a recently recognized, independent prognostic factor in colorectal cancer, but lacks a standardized assessment methodology. Although staining with pan-cytokeratin has been shown to mitigate the issue of lack of reproducible, intra-observer agreement, usage of this antibody remains expensive and is limited in clinical practice. We propose an automated image analysis framework to take advantage of the visual superiority of pan-cytokeratin and the routine use of H&E to detect and quantify tumor budding. Our framework has demonstrated promising ability to identify tumor regions of colorectal slides-92.0% accuracy, 94.5% sensitivity, and 85% specificity-across four independent datasets.

KW - Classification

KW - Colorectal cancer

KW - Deep learning

KW - Pan-cytokeratin

KW - Tumor budding

KW - cGAN

UR - http://www.scopus.com/inward/record.url?scp=85064947106&partnerID=8YFLogxK

U2 - 10.1117/12.2513014

DO - 10.1117/12.2513014

M3 - Conference contribution

AN - SCOPUS:85064947106

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

BT - Medical Imaging 2019

A2 - Tomaszewski, John E.

A2 - Ward, Aaron D.

PB - SPIE

T2 - Medical Imaging 2019: Digital Pathology

Y2 - 20 February 2019 through 21 February 2019

ER -

Colorectal tumor identification by transferring knowledge from pan-cytokeratin to H&E

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this