Automatic detection of small bowel tumors in wireless capsule endoscopy images using ensemble learning
Abstract
Purpose Wireless Capsule Endoscopy (WCE) is a minimally invasive diagnosis tool for lesion detection in the gastrointestinal tract, reaching places where conventional endoscopy is unable to. However, the significant amount of acquired data leads to difficulties in the diagnosis by the physicians; which can be eased with computer assistance. This paper addresses a method for the automatic detection of tumors in WCE by using a two‐step based procedure: region of interest selection and classification. Methods The first step aims to separate abnormal from normal tissue by using automatic segmentation based on a Gaussian Mixture Model (GMM). A modified version of the Anderson method for convergence acceleration of the expectation–maximization (EM) algorithm is proposed. The proposed features for both segmentation and classification are based on the CIELab color space, as a way of bypassing lightness variations, where the L component is discarded. Tissue variability among subjects, light inhomogeneities and even intensity differences among different devices can be overcome by using simultaneously features from both regions. In the second step, an ensemble system with partition of the training data with a new training scheme is proposed. At this stage, the gating network is trained after the experts have been trained decoupling themore »
- Authors:
-
- CMEMS‐UMinho Research Unit University of Minho Guimarães Portugal
- School of Health Polytechnic Institute of Castelo Branco Castelo Branco Portugal
- ALGORITMI Research Center University of Minho Campus of Gualtar 4710‐057 Braga Portugal
- Gastroenterology Department of the Hospital of Braga Braga Portugal, ICVS/3B’s Associate Laboratory University of Minho Braga Portugal
- Publication Date:
- Sponsoring Org.:
- USDOE
- OSTI Identifier:
- 1573850
- Resource Type:
- Publisher's Accepted Manuscript
- Journal Name:
- Medical Physics
- Additional Journal Information:
- Journal Name: Medical Physics Journal Volume: 47 Journal Issue: 1; Journal ID: ISSN 0094-2405
- Publisher:
- Wiley Blackwell (John Wiley & Sons)
- Country of Publication:
- United States
- Language:
- English
Citation Formats
Vieira, Pedro M., Freitas, Nuno R., Valente, João, Vaz, Ismael F., Rolanda, Carla, and Lima, Carlos S. Automatic detection of small bowel tumors in wireless capsule endoscopy images using ensemble learning. United States: N. p., 2019.
Web. doi:10.1002/mp.13709.
Vieira, Pedro M., Freitas, Nuno R., Valente, João, Vaz, Ismael F., Rolanda, Carla, & Lima, Carlos S. Automatic detection of small bowel tumors in wireless capsule endoscopy images using ensemble learning. United States. https://doi.org/10.1002/mp.13709
Vieira, Pedro M., Freitas, Nuno R., Valente, João, Vaz, Ismael F., Rolanda, Carla, and Lima, Carlos S. Mon .
"Automatic detection of small bowel tumors in wireless capsule endoscopy images using ensemble learning". United States. https://doi.org/10.1002/mp.13709.
@article{osti_1573850,
title = {Automatic detection of small bowel tumors in wireless capsule endoscopy images using ensemble learning},
author = {Vieira, Pedro M. and Freitas, Nuno R. and Valente, João and Vaz, Ismael F. and Rolanda, Carla and Lima, Carlos S.},
abstractNote = {Purpose Wireless Capsule Endoscopy (WCE) is a minimally invasive diagnosis tool for lesion detection in the gastrointestinal tract, reaching places where conventional endoscopy is unable to. However, the significant amount of acquired data leads to difficulties in the diagnosis by the physicians; which can be eased with computer assistance. This paper addresses a method for the automatic detection of tumors in WCE by using a two‐step based procedure: region of interest selection and classification. Methods The first step aims to separate abnormal from normal tissue by using automatic segmentation based on a Gaussian Mixture Model (GMM). A modified version of the Anderson method for convergence acceleration of the expectation–maximization (EM) algorithm is proposed. The proposed features for both segmentation and classification are based on the CIELab color space, as a way of bypassing lightness variations, where the L component is discarded. Tissue variability among subjects, light inhomogeneities and even intensity differences among different devices can be overcome by using simultaneously features from both regions. In the second step, an ensemble system with partition of the training data with a new training scheme is proposed. At this stage, the gating network is trained after the experts have been trained decoupling the joint maximization of both modules. The partition module is also used at the test step, leading the incoming data to the most likely expert allowing incremental adaptation by preserving data diversity. Results This algorithm outperforms others based on texture features selected from Wavelets and Curvelets transforms, classified by a regular support vector machine (SVM) in more than 5%. Conclusions This work shows that simpler features can outperform more elaborate ones if appropriately designed. In the current case, luminance was discarded to cope with saturated tissue, facilitating the color perception. Ensemble systems remain an open research field. In the current case, changes in both topology and training strategy have led to significant performance improvements. A system with this level of performance can be used in current clinical practice.},
doi = {10.1002/mp.13709},
journal = {Medical Physics},
number = 1,
volume = 47,
place = {United States},
year = {Mon Nov 11 00:00:00 EST 2019},
month = {Mon Nov 11 00:00:00 EST 2019}
}
https://doi.org/10.1002/mp.13709
Web of Science
Works referenced in this record:
journal, January 2017
- Mahdi, Alizadeh; Omid, Haji; Kaveh, Sharzehi
- The Journal of Biomedical Research, Vol. 31, Issue 5
Small bowel malignancy
journal, January 2004
- de Franchis, Roberto; Rondonotti, Emanuele; Abbiati, Carla
- Gastrointestinal Endoscopy Clinics of North America, Vol. 14, Issue 1
A model of deformable rings for interpretation of wireless capsule endoscopic videos
journal, April 2009
- Szczypiński, Piotr M.; Sriram, Ram D.; Sriram, Parupudi V. J.
- Medical Image Analysis, Vol. 13, Issue 2
A novel least squares support vector machine ensemble model for NOx emission prediction of a coal-fired boiler
journal, June 2013
- Lv, You; Liu, Jizhen; Yang, Tingting
- Energy, Vol. 55
Ensemble of online sequential extreme learning machine
journal, August 2009
- Lan, Yuan; Soh, Yeng Chai; Huang, Guang-Bin
- Neurocomputing, Vol. 72, Issue 13-15
Texture and color based image segmentation and pathology detection in capsule endoscopy videos
journal, January 2014
- Szczypiński, Piotr; Klepaczko, Artur; Pazurek, Marek
- Computer Methods and Programs in Biomedicine, Vol. 113, Issue 1
The usage of soft-computing methodologies in interpreting capsule endoscopy
journal, June 2007
- Kodogiannis, V. S.; Boulougoura, M.; Wadge, E.
- Engineering Applications of Artificial Intelligence, Vol. 20, Issue 4
Small bowel capsule endoscopy: Where are we after almost 15 years of use?
journal, January 2015
- Bruaene, Cedric Van de
- World Journal of Gastrointestinal Endoscopy, Vol. 7, Issue 1
Detection of small bowel tumor based on multi-scale curvelet analysis and fractal technology in capsule endoscopy
journal, March 2016
- Liu, Gang; Yan, Guozheng; Kuang, Shuai
- Computers in Biology and Medicine, Vol. 70
The Usefulness of Capsule Endoscopy for Small Bowel Tumors
journal, January 2016
- Cheung, Dae Young; Kim, Jin Su; Shim, Ki-Nam
- Clinical Endoscopy, Vol. 49, Issue 1
Skin Color Measurements in Terms of CIELAB Color Space Values
journal, October 1992
- Weatherall, Ian L.; Coombs, Bernard D.
- Journal of Investigative Dermatology, Vol. 99, Issue 4
Achieving total enteroscopy with capsule endoscopy in all patients: are we stretching the limits of technology?
journal, January 2009
- Kaffes, Arthur J.
- Gastrointestinal Endoscopy, Vol. 69, Issue 1
Automatic small bowel tumor diagnosis by using multi-scale wavelet-based analysis in wireless capsule endoscopy images
journal, January 2012
- Barbosa, Daniel C.; Roupar, Dalila B.; Ramos, Jaime C.
- BioMedical Engineering OnLine, Vol. 11, Issue 1
Wireless capsule endoscopy
journal, May 2000
- Iddan, Gavriel; Meron, Gavriel; Glukhovsky, Arkady
- Nature, Vol. 405, Issue 6785
Efficient implementation of GLCM based texture feature computation using CUDA platform
conference, May 2017
- Parvez, Asad; Phadke, Anuradha C.
- 2017 International Conference on Trends in Electronics and Informatics (ICOEI), 2017 International Conference on Trends in Electronics and Informatics (ICEI)
Computer-aided tumor detection in endoscopic video using color wavelet features
journal, September 2003
- Karkanis, S. A.; Iakovidis, D. K.; Maroulis, D. E.
- IEEE Transactions on Information Technology in Biomedicine, Vol. 7, Issue 3
Total enteroscopy with a nonsurgical steerable double-balloon method
journal, February 2001
- Yamamoto, Hironori; Sekine, Yutaka; Sato, Yukihiro
- Gastrointestinal Endoscopy, Vol. 53, Issue 2
An efficient implementation of GLCM algorithm in FPGA
conference, December 2018
- Ben Atitallah, M. A.; Kachouri, R.; Kammoun, M.
- 2018 International Conference on Internet of Things, Embedded Systems and Communications (IINTEC)
Anderson Acceleration for Fixed-Point Iterations
journal, January 2011
- Walker, Homer F.; Ni, Peng
- SIAM Journal on Numerical Analysis, Vol. 49, Issue 4
Convergence results for the EM approach to mixtures of experts architectures
journal, January 1995
- Jordan, Michael I.; Xu, Lei
- Neural Networks, Vol. 8, Issue 9
CoLD: a versatile detection system for colorectal lesions in endoscopy video-frames
journal, February 2003
- Maroulis, D. E.; Iakovidis, D. K.; Karkanis, S. A.
- Computer Methods and Programs in Biomedicine, Vol. 70, Issue 2