Multiple classifier systems in texton-based approach for the classification of CT images of Lung

Mehrdad J. Gangeh; Lauge Sørensen; Saher B. Shaker; Mohamed S. Kamel; Marleen de Bruijne

doi:10.1007/978-3-642-18421-5_15

Multiple classifier systems in texton-based approach for the classification of CT images of Lung

Mehrdad J. Gangeh, Lauge Sørensen, Saher B. Shaker, Mohamed S. Kamel, Marleen de Bruijne

Department of Computer Science

7 Citations (Scopus)

Abstract

In this paper, we propose using texton signatures based on raw pixel
representation along with a parallel multiple classifier system for the classification of emphysema in computed tomography images of the lung. The multiple classifier system is composed of support vector machines on the texton signatures as base classifiers and combines their decisions using product rule. The proposed approach is tested on 168 annotated regions of interest consisting of normal tissue, centrilobular emphysema, and paraseptal emphysema. Textonbased approach in texture classification mainly has two parameters, i.e., texton size and k value in k-means. Our results show that while aggregation of single decisions by SVMs over various k values using multiple classifier systems helps to improve the results compared to single SVMs, combining over different
texton sizes is not beneficial. The performance of the proposed system, with an
accuracy of 95%, is similar to a recently proposed approach based on local
binary patterns, which performs almost the best among other approaches in the
literature.

Original language	English
Title of host publication	Medical Computer Vision. Recognition Techniques and Applications in Medical Imaging : International MICCAI Workshop, MCV 2010, Beijing, China, September 20, 2010, Revised Selected Papers
Editors	Bjoern Menze, Georg Langs, Zhowen Tu, Antonio Criminsi
Number of pages	11
Publisher	Springer
Publication date	2011
Pages	153-163
ISBN (Print)	978-3-642-18420-8
ISBN (Electronic)	978-3-642-18421-5
DOIs	https://doi.org/10.1007/978-3-642-18421-5_15
Publication status	Published - 2011
Event	Medical Computer Vision 2010: Recognition Techniques and Applications in Medical Imaging - Beijing, China Duration: 20 Sept 2010 → 20 Sept 2010

Conference

Conference	Medical Computer Vision 2010: Recognition Techniques and Applications in Medical Imaging
Country/Territory	China
City	Beijing
Period	20/09/2010 → 20/09/2010

Series	Lecture notes in computer science
Volume	6533
ISSN	0302-9743

Access to Document

10.1007/978-3-642-18421-5_15

Cite this

Gangeh, M. J., Sørensen, L., Shaker, S. B., Kamel, M. S., & de Bruijne, M. (2011). Multiple classifier systems in texton-based approach for the classification of CT images of Lung. In B. Menze, G. Langs, Z. Tu, & A. Criminsi (Eds.), Medical Computer Vision. Recognition Techniques and Applications in Medical Imaging: International MICCAI Workshop, MCV 2010, Beijing, China, September 20, 2010, Revised Selected Papers (pp. 153-163). Springer. Lecture notes in computer science Vol. 6533 https://doi.org/10.1007/978-3-642-18421-5_15

Multiple classifier systems in texton-based approach for the classification of CT images of Lung. / Gangeh, Mehrdad J.; Sørensen, Lauge; Shaker, Saher B. et al.

Medical Computer Vision. Recognition Techniques and Applications in Medical Imaging: International MICCAI Workshop, MCV 2010, Beijing, China, September 20, 2010, Revised Selected Papers. ed. / Bjoern Menze; Georg Langs; Zhowen Tu; Antonio Criminsi. Springer, 2011. p. 153-163 (Lecture notes in computer science, Vol. 6533).

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

Gangeh, MJ, Sørensen, L, Shaker, SB, Kamel, MS & de Bruijne, M 2011, Multiple classifier systems in texton-based approach for the classification of CT images of Lung. in B Menze, G Langs, Z Tu & A Criminsi (eds), Medical Computer Vision. Recognition Techniques and Applications in Medical Imaging: International MICCAI Workshop, MCV 2010, Beijing, China, September 20, 2010, Revised Selected Papers. Springer, Lecture notes in computer science, vol. 6533, pp. 153-163, Medical Computer Vision 2010: Recognition Techniques and Applications in Medical Imaging, Beijing, China, 20/09/2010. https://doi.org/10.1007/978-3-642-18421-5_15

Gangeh MJ, Sørensen L, Shaker SB, Kamel MS, de Bruijne M. Multiple classifier systems in texton-based approach for the classification of CT images of Lung. In Menze B, Langs G, Tu Z, Criminsi A, editors, Medical Computer Vision. Recognition Techniques and Applications in Medical Imaging: International MICCAI Workshop, MCV 2010, Beijing, China, September 20, 2010, Revised Selected Papers. Springer. 2011. p. 153-163. (Lecture notes in computer science, Vol. 6533). doi: 10.1007/978-3-642-18421-5_15

Gangeh, Mehrdad J. ; Sørensen, Lauge ; Shaker, Saher B. et al. / Multiple classifier systems in texton-based approach for the classification of CT images of Lung. Medical Computer Vision. Recognition Techniques and Applications in Medical Imaging: International MICCAI Workshop, MCV 2010, Beijing, China, September 20, 2010, Revised Selected Papers. editor / Bjoern Menze ; Georg Langs ; Zhowen Tu ; Antonio Criminsi. Springer, 2011. pp. 153-163 (Lecture notes in computer science, Vol. 6533).

@inproceedings{610e7a60902611df928f000ea68e967b,

title = "Multiple classifier systems in texton-based approach for the classification of CT images of Lung",

abstract = "In this paper, we propose using texton signatures based on raw pixelrepresentation along with a parallel multiple classifier system for the classification of emphysema in computed tomography images of the lung. The multiple classifier system is composed of support vector machines on the texton signatures as base classifiers and combines their decisions using product rule. The proposed approach is tested on 168 annotated regions of interest consisting of normal tissue, centrilobular emphysema, and paraseptal emphysema. Textonbased approach in texture classification mainly has two parameters, i.e., texton size and k value in k-means. Our results show that while aggregation of single decisions by SVMs over various k values using multiple classifier systems helps to improve the results compared to single SVMs, combining over differenttexton sizes is not beneficial. The performance of the proposed system, with anaccuracy of 95%, is similar to a recently proposed approach based on localbinary patterns, which performs almost the best among other approaches in theliterature.",

author = "Gangeh, {Mehrdad J.} and Lauge S{\o}rensen and Shaker, {Saher B.} and Kamel, {Mohamed S.} and {de Bruijne}, Marleen",

year = "2011",

doi = "10.1007/978-3-642-18421-5_15",

language = "English",

isbn = "978-3-642-18420-8",

series = "Lecture notes in computer science",

publisher = "Springer",

pages = "153--163",

editor = "Bjoern Menze and Georg Langs and Zhowen Tu and Antonio Criminsi",

booktitle = "Medical Computer Vision. Recognition Techniques and Applications in Medical Imaging",

note = "Medical Computer Vision 2010: Recognition Techniques and Applications in Medical Imaging ; Conference date: 20-09-2010 Through 20-09-2010",

}

TY - GEN

T1 - Multiple classifier systems in texton-based approach for the classification of CT images of Lung

AU - Gangeh, Mehrdad J.

AU - Sørensen, Lauge

AU - Shaker, Saher B.

AU - Kamel, Mohamed S.

AU - de Bruijne, Marleen

PY - 2011

Y1 - 2011

N2 - In this paper, we propose using texton signatures based on raw pixelrepresentation along with a parallel multiple classifier system for the classification of emphysema in computed tomography images of the lung. The multiple classifier system is composed of support vector machines on the texton signatures as base classifiers and combines their decisions using product rule. The proposed approach is tested on 168 annotated regions of interest consisting of normal tissue, centrilobular emphysema, and paraseptal emphysema. Textonbased approach in texture classification mainly has two parameters, i.e., texton size and k value in k-means. Our results show that while aggregation of single decisions by SVMs over various k values using multiple classifier systems helps to improve the results compared to single SVMs, combining over differenttexton sizes is not beneficial. The performance of the proposed system, with anaccuracy of 95%, is similar to a recently proposed approach based on localbinary patterns, which performs almost the best among other approaches in theliterature.

AB - In this paper, we propose using texton signatures based on raw pixelrepresentation along with a parallel multiple classifier system for the classification of emphysema in computed tomography images of the lung. The multiple classifier system is composed of support vector machines on the texton signatures as base classifiers and combines their decisions using product rule. The proposed approach is tested on 168 annotated regions of interest consisting of normal tissue, centrilobular emphysema, and paraseptal emphysema. Textonbased approach in texture classification mainly has two parameters, i.e., texton size and k value in k-means. Our results show that while aggregation of single decisions by SVMs over various k values using multiple classifier systems helps to improve the results compared to single SVMs, combining over differenttexton sizes is not beneficial. The performance of the proposed system, with anaccuracy of 95%, is similar to a recently proposed approach based on localbinary patterns, which performs almost the best among other approaches in theliterature.

U2 - 10.1007/978-3-642-18421-5_15

DO - 10.1007/978-3-642-18421-5_15

M3 - Article in proceedings

SN - 978-3-642-18420-8

T3 - Lecture notes in computer science

SP - 153

EP - 163

BT - Medical Computer Vision. Recognition Techniques and Applications in Medical Imaging

A2 - Menze, Bjoern

A2 - Langs, Georg

A2 - Tu, Zhowen

A2 - Criminsi, Antonio

PB - Springer

T2 - Medical Computer Vision 2010: Recognition Techniques and Applications in Medical Imaging

Y2 - 20 September 2010 through 20 September 2010

ER -

Multiple classifier systems in texton-based approach for the classification of CT images of Lung

Abstract

Conference

Access to Document

Fingerprint

Cite this