Non-informative frame classification in colonoscopy videos using CNNs

A. B.M. Rezbaul Islam; Ali Alammari; Jung Hwan Oh; Wallapak Tavanapong; Johnny Wong; Piet C. de Groen

doi:10.1145/3288200.3288207

Non-informative frame classification in colonoscopy videos using CNNs

A. B.M. Rezbaul Islam, Ali Alammari, Jung Hwan Oh, Wallapak Tavanapong, Johnny Wong, Piet C. de Groen

Medicine - Gastro, Hepatology, Nutrition Division

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

10 Scopus citations

Abstract

In the US, colorectal cancer is the second leading cause of cancer-related deaths behind lung cancer, causing about 49,000 annual deaths. Colonoscopy is currently the gold standard procedure for colorectal cancer screening. However, recent data suggest that there is a significant (4-12%) miss-rate for the detection of even large polyps and cancers. To address this, we have been investigating an „automated feedback system‟ which measures quality of colonoscopy automatically by analyzing colonoscopy video frames in order to assist the endoscopist to improve the quality of the actual procedure being performed. One of the fundamental steps analyzing colonoscopy video frames for the automated quality feedback system is to distinguish non-informative frames from informative ones. Most methods to detect and classify these non-informative frames are based on the hand-engineered features. However, it is very tedious to design optimal hand-engineered features. In this paper, we explore the effectiveness of Convolutional Neural Network (CNN) to detect and classify these non-informative frames. The experimental results show that the proposed approaches are promising.

Original language	English (US)
Title of host publication	ICBSP 2018 - Proceedings of 2018 3rd International Conference on Biomedical Imaging, Signal Processing
Publisher	Association for Computing Machinery
Pages	53-60
Number of pages	8
ISBN (Electronic)	9781450364775
DOIs	https://doi.org/10.1145/3288200.3288207
State	Published - Oct 11 2018
Event	3rd International Conference on Biomedical Imaging, Signal Processing, ICBSP 2018 - Bari, Italy Duration: Oct 11 2018 → Oct 13 2018

Publication series

Name	ACM International Conference Proceeding Series

Other

Other	3rd International Conference on Biomedical Imaging, Signal Processing, ICBSP 2018
Country/Territory	Italy
City	Bari
Period	10/11/18 → 10/13/18

Bibliographical note

Funding Information:
This study was supported in part by a grant from the NIH (Grant #1R01DK106130-01A1). Principal Investigator Piet C. de Groen and Co-Investigators, Oh, Tavanapong, and Wong have equity interest in EndoMetric Corp. Tavanapong, Wong, and Oh hold management positions in EndoMetric Corp. De Groen serves as a Medical advisor of EndoMetric. The terms of this arrangement have been reviewed and approved by Iowa State University in accordance with its conflict of interest policies.

Publisher Copyright:
© 2018 Association for Computing Machinery.

Keywords

Convolutional neural network
Medical image classification
Medical video processing
Ulcerative colitis severity

UN SDGs

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

Publisher link

10.1145/3288200.3288207

Find It

Find It at U of M Libraries

Cite this

Rezbaul Islam, A. B. M., Alammari, A., Oh, J. H., Tavanapong, W., Wong, J., & de Groen, P. C. (2018). Non-informative frame classification in colonoscopy videos using CNNs. In ICBSP 2018 - Proceedings of 2018 3rd International Conference on Biomedical Imaging, Signal Processing (pp. 53-60). (ACM International Conference Proceeding Series). Association for Computing Machinery. https://doi.org/10.1145/3288200.3288207

Non-informative frame classification in colonoscopy videos using CNNs. / Rezbaul Islam, A. B.M.; Alammari, Ali; Oh, Jung Hwan et al.
ICBSP 2018 - Proceedings of 2018 3rd International Conference on Biomedical Imaging, Signal Processing. Association for Computing Machinery, 2018. p. 53-60 (ACM International Conference Proceeding Series).

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

Rezbaul Islam, ABM, Alammari, A, Oh, JH, Tavanapong, W, Wong, J & de Groen, PC 2018, Non-informative frame classification in colonoscopy videos using CNNs. in ICBSP 2018 - Proceedings of 2018 3rd International Conference on Biomedical Imaging, Signal Processing. ACM International Conference Proceeding Series, Association for Computing Machinery, pp. 53-60, 3rd International Conference on Biomedical Imaging, Signal Processing, ICBSP 2018, Bari, Italy, 10/11/18. https://doi.org/10.1145/3288200.3288207

Rezbaul Islam ABM, Alammari A, Oh JH, Tavanapong W, Wong J, de Groen PC. Non-informative frame classification in colonoscopy videos using CNNs. In ICBSP 2018 - Proceedings of 2018 3rd International Conference on Biomedical Imaging, Signal Processing. Association for Computing Machinery. 2018. p. 53-60. (ACM International Conference Proceeding Series). doi: 10.1145/3288200.3288207

@inproceedings{b62e0d167c244f17b0a30da9063f9453,

title = "Non-informative frame classification in colonoscopy videos using CNNs",

abstract = "In the US, colorectal cancer is the second leading cause of cancer-related deaths behind lung cancer, causing about 49,000 annual deaths. Colonoscopy is currently the gold standard procedure for colorectal cancer screening. However, recent data suggest that there is a significant (4-12%) miss-rate for the detection of even large polyps and cancers. To address this, we have been investigating an „automated feedback system‟ which measures quality of colonoscopy automatically by analyzing colonoscopy video frames in order to assist the endoscopist to improve the quality of the actual procedure being performed. One of the fundamental steps analyzing colonoscopy video frames for the automated quality feedback system is to distinguish non-informative frames from informative ones. Most methods to detect and classify these non-informative frames are based on the hand-engineered features. However, it is very tedious to design optimal hand-engineered features. In this paper, we explore the effectiveness of Convolutional Neural Network (CNN) to detect and classify these non-informative frames. The experimental results show that the proposed approaches are promising.",

keywords = "Convolutional neural network, Medical image classification, Medical video processing, Ulcerative colitis severity",

author = "{Rezbaul Islam}, {A. B.M.} and Ali Alammari and Oh, {Jung Hwan} and Wallapak Tavanapong and Johnny Wong and {de Groen}, {Piet C.}",

note = "Funding Information: This study was supported in part by a grant from the NIH (Grant #1R01DK106130-01A1). Principal Investigator Piet C. de Groen and Co-Investigators, Oh, Tavanapong, and Wong have equity interest in EndoMetric Corp. Tavanapong, Wong, and Oh hold management positions in EndoMetric Corp. De Groen serves as a Medical advisor of EndoMetric. The terms of this arrangement have been reviewed and approved by Iowa State University in accordance with its conflict of interest policies. Publisher Copyright: {\textcopyright} 2018 Association for Computing Machinery.; 3rd International Conference on Biomedical Imaging, Signal Processing, ICBSP 2018 ; Conference date: 11-10-2018 Through 13-10-2018",

year = "2018",

month = oct,

day = "11",

doi = "10.1145/3288200.3288207",

language = "English (US)",

series = "ACM International Conference Proceeding Series",

publisher = "Association for Computing Machinery",

pages = "53--60",

booktitle = "ICBSP 2018 - Proceedings of 2018 3rd International Conference on Biomedical Imaging, Signal Processing",

}

TY - GEN

T1 - Non-informative frame classification in colonoscopy videos using CNNs

AU - Rezbaul Islam, A. B.M.

AU - Alammari, Ali

AU - Oh, Jung Hwan

AU - Tavanapong, Wallapak

AU - Wong, Johnny

AU - de Groen, Piet C.

N1 - Funding Information: This study was supported in part by a grant from the NIH (Grant #1R01DK106130-01A1). Principal Investigator Piet C. de Groen and Co-Investigators, Oh, Tavanapong, and Wong have equity interest in EndoMetric Corp. Tavanapong, Wong, and Oh hold management positions in EndoMetric Corp. De Groen serves as a Medical advisor of EndoMetric. The terms of this arrangement have been reviewed and approved by Iowa State University in accordance with its conflict of interest policies. Publisher Copyright: © 2018 Association for Computing Machinery.

PY - 2018/10/11

Y1 - 2018/10/11

N2 - In the US, colorectal cancer is the second leading cause of cancer-related deaths behind lung cancer, causing about 49,000 annual deaths. Colonoscopy is currently the gold standard procedure for colorectal cancer screening. However, recent data suggest that there is a significant (4-12%) miss-rate for the detection of even large polyps and cancers. To address this, we have been investigating an „automated feedback system‟ which measures quality of colonoscopy automatically by analyzing colonoscopy video frames in order to assist the endoscopist to improve the quality of the actual procedure being performed. One of the fundamental steps analyzing colonoscopy video frames for the automated quality feedback system is to distinguish non-informative frames from informative ones. Most methods to detect and classify these non-informative frames are based on the hand-engineered features. However, it is very tedious to design optimal hand-engineered features. In this paper, we explore the effectiveness of Convolutional Neural Network (CNN) to detect and classify these non-informative frames. The experimental results show that the proposed approaches are promising.

AB - In the US, colorectal cancer is the second leading cause of cancer-related deaths behind lung cancer, causing about 49,000 annual deaths. Colonoscopy is currently the gold standard procedure for colorectal cancer screening. However, recent data suggest that there is a significant (4-12%) miss-rate for the detection of even large polyps and cancers. To address this, we have been investigating an „automated feedback system‟ which measures quality of colonoscopy automatically by analyzing colonoscopy video frames in order to assist the endoscopist to improve the quality of the actual procedure being performed. One of the fundamental steps analyzing colonoscopy video frames for the automated quality feedback system is to distinguish non-informative frames from informative ones. Most methods to detect and classify these non-informative frames are based on the hand-engineered features. However, it is very tedious to design optimal hand-engineered features. In this paper, we explore the effectiveness of Convolutional Neural Network (CNN) to detect and classify these non-informative frames. The experimental results show that the proposed approaches are promising.

KW - Convolutional neural network

KW - Medical image classification

KW - Medical video processing

KW - Ulcerative colitis severity

UR - http://www.scopus.com/inward/record.url?scp=85061053749&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85061053749&partnerID=8YFLogxK

U2 - 10.1145/3288200.3288207

DO - 10.1145/3288200.3288207

M3 - Conference contribution

AN - SCOPUS:85061053749

T3 - ACM International Conference Proceeding Series

SP - 53

EP - 60

BT - ICBSP 2018 - Proceedings of 2018 3rd International Conference on Biomedical Imaging, Signal Processing

PB - Association for Computing Machinery

T2 - 3rd International Conference on Biomedical Imaging, Signal Processing, ICBSP 2018

Y2 - 11 October 2018 through 13 October 2018

ER -

Non-informative frame classification in colonoscopy videos using CNNs

Abstract

Publication series

Other

Bibliographical note

Keywords

UN SDGs

Publisher link

Find It

Other files and links

Fingerprint

Cite this