Dysplasia Grading of Colorectal Polyps Through Convolutional Neural Network Analysis of Whole Slide Images

Daniele Perlo; Enzo Tartaglione; Luca Bertero; Paola Cassoni; Marco Grangetto

doi:10.1007/978-981-16-3880-0_34

Dysplasia Grading of Colorectal Polyps Through Convolutional Neural Network Analysis of Whole Slide Images

Daniele Perlo^*, Enzo Tartaglione, Luca Bertero, Paola Cassoni, Marco Grangetto

^*Corresponding author for this work

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

3 Citations (Scopus)

Abstract

Colorectal cancer is a leading cause of cancer death for both men and women. For this reason, histo-pathological characterization of colorectal polyps is the major instrument for the pathologist in order to infer the actual risk for cancer and to guide further follow-up. Colorectal polyps diagnosis includes the evaluation of the polyp type, and more importantly, the grade of dysplasia. This latter evaluation represents a critical step for the clinical follow-up. The proposed deep learning-based classification pipeline is based on state-of-the-art convolutional neural network, trained using proper countermeasures to tackle WSI high resolution and very imbalanced dataset. The experimental results show that one can successfully classify adenomas dysplasia grade with 70% accuracy, which is in line with the pathologists’ concordance.

Original language	English
Title of host publication	Proceedings of 2021 International Conference on Medical Imaging and Computer-Aided Diagnosis, MICAD 2021 - Medical Imaging and Computer-Aided Diagnosis
Editors	Ruidan Su, Yu-Dong Zhang, Han Liu
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	325-334
Number of pages	10
ISBN (Print)	9789811638794
DOIs	https://doi.org/10.1007/978-981-16-3880-0_34
Publication status	Published - 2022
Externally published	Yes
Event	International Conference on Medical Imaging and Computer-Aided Diagnosis, MICAD 2021 - Virtual, Online Duration: 25 Mar 2021 → 26 Mar 2021

Publication series

Name	Lecture Notes in Electrical Engineering
Volume	784 LNEE
ISSN (Print)	1876-1100
ISSN (Electronic)	1876-1119

Conference

Conference	International Conference on Medical Imaging and Computer-Aided Diagnosis, MICAD 2021
City	Virtual, Online
Period	25/03/21 → 26/03/21

Keywords

Colorectal adenomas
Colorectal polyps
Deep learning
Digital pathology
Multi resolution

Access to Document

10.1007/978-981-16-3880-0_34

Cite this

Perlo, D., Tartaglione, E., Bertero, L., Cassoni, P., & Grangetto, M. (2022). Dysplasia Grading of Colorectal Polyps Through Convolutional Neural Network Analysis of Whole Slide Images. In R. Su, Y.-D. Zhang, & H. Liu (Eds.), Proceedings of 2021 International Conference on Medical Imaging and Computer-Aided Diagnosis, MICAD 2021 - Medical Imaging and Computer-Aided Diagnosis (pp. 325-334). (Lecture Notes in Electrical Engineering; Vol. 784 LNEE). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-981-16-3880-0_34

Perlo, Daniele ; Tartaglione, Enzo ; Bertero, Luca et al. / Dysplasia Grading of Colorectal Polyps Through Convolutional Neural Network Analysis of Whole Slide Images. Proceedings of 2021 International Conference on Medical Imaging and Computer-Aided Diagnosis, MICAD 2021 - Medical Imaging and Computer-Aided Diagnosis. editor / Ruidan Su ; Yu-Dong Zhang ; Han Liu. Springer Science and Business Media Deutschland GmbH, 2022. pp. 325-334 (Lecture Notes in Electrical Engineering).

@inproceedings{65a7804a44dd4fa78586d825b18268da,

title = "Dysplasia Grading of Colorectal Polyps Through Convolutional Neural Network Analysis of Whole Slide Images",

abstract = "Colorectal cancer is a leading cause of cancer death for both men and women. For this reason, histo-pathological characterization of colorectal polyps is the major instrument for the pathologist in order to infer the actual risk for cancer and to guide further follow-up. Colorectal polyps diagnosis includes the evaluation of the polyp type, and more importantly, the grade of dysplasia. This latter evaluation represents a critical step for the clinical follow-up. The proposed deep learning-based classification pipeline is based on state-of-the-art convolutional neural network, trained using proper countermeasures to tackle WSI high resolution and very imbalanced dataset. The experimental results show that one can successfully classify adenomas dysplasia grade with 70% accuracy, which is in line with the pathologists{\textquoteright} concordance.",

keywords = "Colorectal adenomas, Colorectal polyps, Deep learning, Digital pathology, Multi resolution",

author = "Daniele Perlo and Enzo Tartaglione and Luca Bertero and Paola Cassoni and Marco Grangetto",

note = "Publisher Copyright: {\textcopyright} 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.; International Conference on Medical Imaging and Computer-Aided Diagnosis, MICAD 2021 ; Conference date: 25-03-2021 Through 26-03-2021",

year = "2022",

doi = "10.1007/978-981-16-3880-0_34",

language = "English",

isbn = "9789811638794",

series = "Lecture Notes in Electrical Engineering",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "325--334",

editor = "Ruidan Su and Yu-Dong Zhang and Han Liu",

booktitle = "Proceedings of 2021 International Conference on Medical Imaging and Computer-Aided Diagnosis, MICAD 2021 - Medical Imaging and Computer-Aided Diagnosis",

address = "Germany",

}

Perlo, D, Tartaglione, E, Bertero, L, Cassoni, P & Grangetto, M 2022, Dysplasia Grading of Colorectal Polyps Through Convolutional Neural Network Analysis of Whole Slide Images. in R Su, Y-D Zhang & H Liu (eds), Proceedings of 2021 International Conference on Medical Imaging and Computer-Aided Diagnosis, MICAD 2021 - Medical Imaging and Computer-Aided Diagnosis. Lecture Notes in Electrical Engineering, vol. 784 LNEE, Springer Science and Business Media Deutschland GmbH, pp. 325-334, International Conference on Medical Imaging and Computer-Aided Diagnosis, MICAD 2021, Virtual, Online, 25/03/21. https://doi.org/10.1007/978-981-16-3880-0_34

Dysplasia Grading of Colorectal Polyps Through Convolutional Neural Network Analysis of Whole Slide Images. / Perlo, Daniele; Tartaglione, Enzo; Bertero, Luca et al.
Proceedings of 2021 International Conference on Medical Imaging and Computer-Aided Diagnosis, MICAD 2021 - Medical Imaging and Computer-Aided Diagnosis. ed. / Ruidan Su; Yu-Dong Zhang; Han Liu. Springer Science and Business Media Deutschland GmbH, 2022. p. 325-334 (Lecture Notes in Electrical Engineering; Vol. 784 LNEE).

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

TY - GEN

T1 - Dysplasia Grading of Colorectal Polyps Through Convolutional Neural Network Analysis of Whole Slide Images

AU - Perlo, Daniele

AU - Tartaglione, Enzo

AU - Bertero, Luca

AU - Cassoni, Paola

AU - Grangetto, Marco

PY - 2022

Y1 - 2022

N2 - Colorectal cancer is a leading cause of cancer death for both men and women. For this reason, histo-pathological characterization of colorectal polyps is the major instrument for the pathologist in order to infer the actual risk for cancer and to guide further follow-up. Colorectal polyps diagnosis includes the evaluation of the polyp type, and more importantly, the grade of dysplasia. This latter evaluation represents a critical step for the clinical follow-up. The proposed deep learning-based classification pipeline is based on state-of-the-art convolutional neural network, trained using proper countermeasures to tackle WSI high resolution and very imbalanced dataset. The experimental results show that one can successfully classify adenomas dysplasia grade with 70% accuracy, which is in line with the pathologists’ concordance.

AB - Colorectal cancer is a leading cause of cancer death for both men and women. For this reason, histo-pathological characterization of colorectal polyps is the major instrument for the pathologist in order to infer the actual risk for cancer and to guide further follow-up. Colorectal polyps diagnosis includes the evaluation of the polyp type, and more importantly, the grade of dysplasia. This latter evaluation represents a critical step for the clinical follow-up. The proposed deep learning-based classification pipeline is based on state-of-the-art convolutional neural network, trained using proper countermeasures to tackle WSI high resolution and very imbalanced dataset. The experimental results show that one can successfully classify adenomas dysplasia grade with 70% accuracy, which is in line with the pathologists’ concordance.

KW - Colorectal adenomas

KW - Colorectal polyps

KW - Deep learning

KW - Digital pathology

KW - Multi resolution

UR - http://www.scopus.com/inward/record.url?scp=85115167864&partnerID=8YFLogxK

U2 - 10.1007/978-981-16-3880-0_34

DO - 10.1007/978-981-16-3880-0_34

M3 - Conference contribution

AN - SCOPUS:85115167864

SN - 9789811638794

T3 - Lecture Notes in Electrical Engineering

SP - 325

EP - 334

BT - Proceedings of 2021 International Conference on Medical Imaging and Computer-Aided Diagnosis, MICAD 2021 - Medical Imaging and Computer-Aided Diagnosis

A2 - Su, Ruidan

A2 - Zhang, Yu-Dong

A2 - Liu, Han

PB - Springer Science and Business Media Deutschland GmbH

T2 - International Conference on Medical Imaging and Computer-Aided Diagnosis, MICAD 2021

Y2 - 25 March 2021 through 26 March 2021

ER -

Perlo D, Tartaglione E, Bertero L, Cassoni P, Grangetto M. Dysplasia Grading of Colorectal Polyps Through Convolutional Neural Network Analysis of Whole Slide Images. In Su R, Zhang YD, Liu H, editors, Proceedings of 2021 International Conference on Medical Imaging and Computer-Aided Diagnosis, MICAD 2021 - Medical Imaging and Computer-Aided Diagnosis. Springer Science and Business Media Deutschland GmbH. 2022. p. 325-334. (Lecture Notes in Electrical Engineering). doi: 10.1007/978-981-16-3880-0_34

Dysplasia Grading of Colorectal Polyps Through Convolutional Neural Network Analysis of Whole Slide Images

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this