A Benchmark for Endoluminal Scene Segmentation of Colonoscopy Images
Colorectal cancer (CRC) is the third cause of cancer death worldwide. Currently, the standard approach to reduce CRC-related mortality is to perform regular screening in search for polyps and colonoscopy is the screening tool of choice. The main limitations of this screening procedure are polyp miss...
Main Authors: | , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Hindawi Limited
2017-01-01
|
Series: | Journal of Healthcare Engineering |
Online Access: | http://dx.doi.org/10.1155/2017/4037190 |
id |
doaj-d17a9314aef34fd18a35e9d672f92ca0 |
---|---|
record_format |
Article |
spelling |
doaj-d17a9314aef34fd18a35e9d672f92ca02020-11-24T23:14:26ZengHindawi LimitedJournal of Healthcare Engineering2040-22952040-23092017-01-01201710.1155/2017/40371904037190A Benchmark for Endoluminal Scene Segmentation of Colonoscopy ImagesDavid Vázquez0Jorge Bernal1F. Javier Sánchez2Gloria Fernández-Esparrach3Antonio M. López4Adriana Romero5Michal Drozdzal6Aaron Courville7Computer Vision Center, Computer Science Department, Universitat Autonoma de Barcelona, Barcelona, SpainComputer Vision Center, Computer Science Department, Universitat Autonoma de Barcelona, Barcelona, SpainComputer Vision Center, Computer Science Department, Universitat Autonoma de Barcelona, Barcelona, SpainEndoscopy Unit, Gastroenterology Service, CIBERHED, IDIBAPS, Hospital Clinic, Universidad de Barcelona, Barcelona, SpainComputer Vision Center, Computer Science Department, Universitat Autonoma de Barcelona, Barcelona, SpainMontreal Institute for Learning Algorithms, Université de Montréal, Montreal, QC, CanadaÉcole Polytechnique de Montréal, Montréal, QC, CanadaMontreal Institute for Learning Algorithms, Université de Montréal, Montreal, QC, CanadaColorectal cancer (CRC) is the third cause of cancer death worldwide. Currently, the standard approach to reduce CRC-related mortality is to perform regular screening in search for polyps and colonoscopy is the screening tool of choice. The main limitations of this screening procedure are polyp miss rate and the inability to perform visual assessment of polyp malignancy. These drawbacks can be reduced by designing decision support systems (DSS) aiming to help clinicians in the different stages of the procedure by providing endoluminal scene segmentation. Thus, in this paper, we introduce an extended benchmark of colonoscopy image segmentation, with the hope of establishing a new strong benchmark for colonoscopy image analysis research. The proposed dataset consists of 4 relevant classes to inspect the endoluminal scene, targeting different clinical needs. Together with the dataset and taking advantage of advances in semantic segmentation literature, we provide new baselines by training standard fully convolutional networks (FCNs). We perform a comparative study to show that FCNs significantly outperform, without any further postprocessing, prior results in endoluminal scene segmentation, especially with respect to polyp segmentation and localization.http://dx.doi.org/10.1155/2017/4037190 |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
David Vázquez Jorge Bernal F. Javier Sánchez Gloria Fernández-Esparrach Antonio M. López Adriana Romero Michal Drozdzal Aaron Courville |
spellingShingle |
David Vázquez Jorge Bernal F. Javier Sánchez Gloria Fernández-Esparrach Antonio M. López Adriana Romero Michal Drozdzal Aaron Courville A Benchmark for Endoluminal Scene Segmentation of Colonoscopy Images Journal of Healthcare Engineering |
author_facet |
David Vázquez Jorge Bernal F. Javier Sánchez Gloria Fernández-Esparrach Antonio M. López Adriana Romero Michal Drozdzal Aaron Courville |
author_sort |
David Vázquez |
title |
A Benchmark for Endoluminal Scene Segmentation of Colonoscopy Images |
title_short |
A Benchmark for Endoluminal Scene Segmentation of Colonoscopy Images |
title_full |
A Benchmark for Endoluminal Scene Segmentation of Colonoscopy Images |
title_fullStr |
A Benchmark for Endoluminal Scene Segmentation of Colonoscopy Images |
title_full_unstemmed |
A Benchmark for Endoluminal Scene Segmentation of Colonoscopy Images |
title_sort |
benchmark for endoluminal scene segmentation of colonoscopy images |
publisher |
Hindawi Limited |
series |
Journal of Healthcare Engineering |
issn |
2040-2295 2040-2309 |
publishDate |
2017-01-01 |
description |
Colorectal cancer (CRC) is the third cause of cancer death worldwide. Currently, the standard approach to reduce CRC-related mortality is to perform regular screening in search for polyps and colonoscopy is the screening tool of choice. The main limitations of this screening procedure are polyp miss rate and the inability to perform visual assessment of polyp malignancy. These drawbacks can be reduced by designing decision support systems (DSS) aiming to help clinicians in the different stages of the procedure by providing endoluminal scene segmentation. Thus, in this paper, we introduce an extended benchmark of colonoscopy image segmentation, with the hope of establishing a new strong benchmark for colonoscopy image analysis research. The proposed dataset consists of 4 relevant classes to inspect the endoluminal scene, targeting different clinical needs. Together with the dataset and taking advantage of advances in semantic segmentation literature, we provide new baselines by training standard fully convolutional networks (FCNs). We perform a comparative study to show that FCNs significantly outperform, without any further postprocessing, prior results in endoluminal scene segmentation, especially with respect to polyp segmentation and localization. |
url |
http://dx.doi.org/10.1155/2017/4037190 |
work_keys_str_mv |
AT davidvazquez abenchmarkforendoluminalscenesegmentationofcolonoscopyimages AT jorgebernal abenchmarkforendoluminalscenesegmentationofcolonoscopyimages AT fjaviersanchez abenchmarkforendoluminalscenesegmentationofcolonoscopyimages AT gloriafernandezesparrach abenchmarkforendoluminalscenesegmentationofcolonoscopyimages AT antoniomlopez abenchmarkforendoluminalscenesegmentationofcolonoscopyimages AT adrianaromero abenchmarkforendoluminalscenesegmentationofcolonoscopyimages AT michaldrozdzal abenchmarkforendoluminalscenesegmentationofcolonoscopyimages AT aaroncourville abenchmarkforendoluminalscenesegmentationofcolonoscopyimages AT davidvazquez benchmarkforendoluminalscenesegmentationofcolonoscopyimages AT jorgebernal benchmarkforendoluminalscenesegmentationofcolonoscopyimages AT fjaviersanchez benchmarkforendoluminalscenesegmentationofcolonoscopyimages AT gloriafernandezesparrach benchmarkforendoluminalscenesegmentationofcolonoscopyimages AT antoniomlopez benchmarkforendoluminalscenesegmentationofcolonoscopyimages AT adrianaromero benchmarkforendoluminalscenesegmentationofcolonoscopyimages AT michaldrozdzal benchmarkforendoluminalscenesegmentationofcolonoscopyimages AT aaroncourville benchmarkforendoluminalscenesegmentationofcolonoscopyimages |
_version_ |
1725594334104911872 |