Automatic polyp detection in colonoscopy videos

Zijie Yuan, Mohammadhassan Izadyyazdanabadi, Divya Mokkapati, Rujuta Panvalkar, Jae Y. Shin, Nima Tajbakhsh, Suryakanth Gurudu, Jianming Liang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Colon cancer is the second cancer killer in the US [1]. Colonoscopy is the primary method for screening and prevention of colon cancer, but during colonoscopy, a significant number (25% [2]) of polyps (precancerous abnormal growths inside of the colon) are missed; therefore, the goal of our research is to reduce the polyp miss-rate of colonoscopy. This paper presents a method to detect polyp automatically in a colonoscopy video. Our system has two stages: Candidate generation and candidate classification. In candidate generation (stage 1), we chose 3,463 frames (including 1,718 with-polyp frames) from real-time colonoscopy video database. We first applied processing procedures, namely intensity adjustment, edge detection and morphology operations, as pre-preparation. We extracted each connected component (edge contour) as one candidate patch from the pre-processed image. With the help of ground truth (GT) images, 2 constraints were implemented on each candidate patch, dividing and saving them into polyp group and non-polyp group. In candidate classification (stage 2), we trained and tested convolutional neural networks (CNNs) with AlexNet architecture [3] to classify each candidate into with-polyp or non-polyp class. Each with-polyp patch was processed by rotation, translation and scaling for invariant to get a much robust CNNs system. We applied leave-2-patients-out cross-validation on this model (4 of 6 cases were chosen as training set and the rest 2 were as testing set). The system accuracy and sensitivity are 91.47% and 91.76%, respectively.

Original languageEnglish (US)
Title of host publicationMedical Imaging 2017
Subtitle of host publicationImage Processing
PublisherSPIE
Volume10133
ISBN (Electronic)9781510607118
DOIs
StatePublished - 2017
EventMedical Imaging 2017: Image Processing - Orlando, United States
Duration: Feb 12 2017Feb 14 2017

Other

OtherMedical Imaging 2017: Image Processing
CountryUnited States
CityOrlando
Period2/12/172/14/17

Fingerprint

Colonoscopy
Polyps
cancer
Neural networks
Edge detection
Screening
ground truth
edge detection
Testing
Colonic Neoplasms
education
screening
Processing
adjusting
scaling
preparation
sensitivity
Second Primary Neoplasms
Colon
Databases

Keywords

  • Colonoscopy
  • Computer-aided detection
  • Convolutional neural networks (CNNs)
  • Deep learning
  • Polyp

ASJC Scopus subject areas

  • Atomic and Molecular Physics, and Optics
  • Electronic, Optical and Magnetic Materials
  • Biomaterials
  • Radiology Nuclear Medicine and imaging

Cite this

Yuan, Z., Izadyyazdanabadi, M., Mokkapati, D., Panvalkar, R., Shin, J. Y., Tajbakhsh, N., ... Liang, J. (2017). Automatic polyp detection in colonoscopy videos. In Medical Imaging 2017: Image Processing (Vol. 10133). [101332K] SPIE. https://doi.org/10.1117/12.2254671

Automatic polyp detection in colonoscopy videos. / Yuan, Zijie; Izadyyazdanabadi, Mohammadhassan; Mokkapati, Divya; Panvalkar, Rujuta; Shin, Jae Y.; Tajbakhsh, Nima; Gurudu, Suryakanth; Liang, Jianming.

Medical Imaging 2017: Image Processing. Vol. 10133 SPIE, 2017. 101332K.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Yuan, Z, Izadyyazdanabadi, M, Mokkapati, D, Panvalkar, R, Shin, JY, Tajbakhsh, N, Gurudu, S & Liang, J 2017, Automatic polyp detection in colonoscopy videos. in Medical Imaging 2017: Image Processing. vol. 10133, 101332K, SPIE, Medical Imaging 2017: Image Processing, Orlando, United States, 2/12/17. https://doi.org/10.1117/12.2254671
Yuan Z, Izadyyazdanabadi M, Mokkapati D, Panvalkar R, Shin JY, Tajbakhsh N et al. Automatic polyp detection in colonoscopy videos. In Medical Imaging 2017: Image Processing. Vol. 10133. SPIE. 2017. 101332K https://doi.org/10.1117/12.2254671
Yuan, Zijie ; Izadyyazdanabadi, Mohammadhassan ; Mokkapati, Divya ; Panvalkar, Rujuta ; Shin, Jae Y. ; Tajbakhsh, Nima ; Gurudu, Suryakanth ; Liang, Jianming. / Automatic polyp detection in colonoscopy videos. Medical Imaging 2017: Image Processing. Vol. 10133 SPIE, 2017.
@inproceedings{aeed2663e2d1407f9a5ffdbe346f739d,
title = "Automatic polyp detection in colonoscopy videos",
abstract = "Colon cancer is the second cancer killer in the US [1]. Colonoscopy is the primary method for screening and prevention of colon cancer, but during colonoscopy, a significant number (25{\%} [2]) of polyps (precancerous abnormal growths inside of the colon) are missed; therefore, the goal of our research is to reduce the polyp miss-rate of colonoscopy. This paper presents a method to detect polyp automatically in a colonoscopy video. Our system has two stages: Candidate generation and candidate classification. In candidate generation (stage 1), we chose 3,463 frames (including 1,718 with-polyp frames) from real-time colonoscopy video database. We first applied processing procedures, namely intensity adjustment, edge detection and morphology operations, as pre-preparation. We extracted each connected component (edge contour) as one candidate patch from the pre-processed image. With the help of ground truth (GT) images, 2 constraints were implemented on each candidate patch, dividing and saving them into polyp group and non-polyp group. In candidate classification (stage 2), we trained and tested convolutional neural networks (CNNs) with AlexNet architecture [3] to classify each candidate into with-polyp or non-polyp class. Each with-polyp patch was processed by rotation, translation and scaling for invariant to get a much robust CNNs system. We applied leave-2-patients-out cross-validation on this model (4 of 6 cases were chosen as training set and the rest 2 were as testing set). The system accuracy and sensitivity are 91.47{\%} and 91.76{\%}, respectively.",
keywords = "Colonoscopy, Computer-aided detection, Convolutional neural networks (CNNs), Deep learning, Polyp",
author = "Zijie Yuan and Mohammadhassan Izadyyazdanabadi and Divya Mokkapati and Rujuta Panvalkar and Shin, {Jae Y.} and Nima Tajbakhsh and Suryakanth Gurudu and Jianming Liang",
year = "2017",
doi = "10.1117/12.2254671",
language = "English (US)",
volume = "10133",
booktitle = "Medical Imaging 2017",
publisher = "SPIE",

}

TY - GEN

T1 - Automatic polyp detection in colonoscopy videos

AU - Yuan, Zijie

AU - Izadyyazdanabadi, Mohammadhassan

AU - Mokkapati, Divya

AU - Panvalkar, Rujuta

AU - Shin, Jae Y.

AU - Tajbakhsh, Nima

AU - Gurudu, Suryakanth

AU - Liang, Jianming

PY - 2017

Y1 - 2017

N2 - Colon cancer is the second cancer killer in the US [1]. Colonoscopy is the primary method for screening and prevention of colon cancer, but during colonoscopy, a significant number (25% [2]) of polyps (precancerous abnormal growths inside of the colon) are missed; therefore, the goal of our research is to reduce the polyp miss-rate of colonoscopy. This paper presents a method to detect polyp automatically in a colonoscopy video. Our system has two stages: Candidate generation and candidate classification. In candidate generation (stage 1), we chose 3,463 frames (including 1,718 with-polyp frames) from real-time colonoscopy video database. We first applied processing procedures, namely intensity adjustment, edge detection and morphology operations, as pre-preparation. We extracted each connected component (edge contour) as one candidate patch from the pre-processed image. With the help of ground truth (GT) images, 2 constraints were implemented on each candidate patch, dividing and saving them into polyp group and non-polyp group. In candidate classification (stage 2), we trained and tested convolutional neural networks (CNNs) with AlexNet architecture [3] to classify each candidate into with-polyp or non-polyp class. Each with-polyp patch was processed by rotation, translation and scaling for invariant to get a much robust CNNs system. We applied leave-2-patients-out cross-validation on this model (4 of 6 cases were chosen as training set and the rest 2 were as testing set). The system accuracy and sensitivity are 91.47% and 91.76%, respectively.

AB - Colon cancer is the second cancer killer in the US [1]. Colonoscopy is the primary method for screening and prevention of colon cancer, but during colonoscopy, a significant number (25% [2]) of polyps (precancerous abnormal growths inside of the colon) are missed; therefore, the goal of our research is to reduce the polyp miss-rate of colonoscopy. This paper presents a method to detect polyp automatically in a colonoscopy video. Our system has two stages: Candidate generation and candidate classification. In candidate generation (stage 1), we chose 3,463 frames (including 1,718 with-polyp frames) from real-time colonoscopy video database. We first applied processing procedures, namely intensity adjustment, edge detection and morphology operations, as pre-preparation. We extracted each connected component (edge contour) as one candidate patch from the pre-processed image. With the help of ground truth (GT) images, 2 constraints were implemented on each candidate patch, dividing and saving them into polyp group and non-polyp group. In candidate classification (stage 2), we trained and tested convolutional neural networks (CNNs) with AlexNet architecture [3] to classify each candidate into with-polyp or non-polyp class. Each with-polyp patch was processed by rotation, translation and scaling for invariant to get a much robust CNNs system. We applied leave-2-patients-out cross-validation on this model (4 of 6 cases were chosen as training set and the rest 2 were as testing set). The system accuracy and sensitivity are 91.47% and 91.76%, respectively.

KW - Colonoscopy

KW - Computer-aided detection

KW - Convolutional neural networks (CNNs)

KW - Deep learning

KW - Polyp

UR - http://www.scopus.com/inward/record.url?scp=85020297930&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85020297930&partnerID=8YFLogxK

U2 - 10.1117/12.2254671

DO - 10.1117/12.2254671

M3 - Conference contribution

VL - 10133

BT - Medical Imaging 2017

PB - SPIE

ER -