Classification of colorectal polyp regions in optical projection tomography

Wenqi Li; Jianguo Zhang; Stephen J. McKenna; Maria Coats; Frank A. Carey

doi:10.1109/ISBI.2013.6556580

Classification of colorectal polyp regions in optical projection tomography

Wenqi Li, Jianguo Zhang, Stephen J. McKenna, Maria Coats, Frank A. Carey

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

6 Citations (Scopus)

Abstract

The potential of optical projection tomography (OPT) to enhance colorectal polyp diagnosis is beginning to be explored. This paper presents, to the best of our knowledge, the first study on automatic image analysis of OPT images of colorectal polyps. 3D regions are classified using the bag of visual words framework and support vector machines. Independent subspace analysis is used to learn a domain-specific feature dictionary. This is compared to the use of raw patches (after random projection) and local binary patterns. Classification experiments (across patients) at the patch level and at the region level are presented using a set of 30 expert-annotated OPT images. Results show that accurate classification of 3D OPT image regions is feasible using this approach; regions of low-grade dysplasia and invasive cancer were discriminated with approximately 90% accuracy.

Original language	English
Title of host publication	ISBI 2013
Subtitle of host publication	IEEE 10th International Symposium on Biomedical Imaging: From Nano to Macro - Proceedings
Place of Publication	Piscataway, N.J.
Publisher	IEEE
Pages	736-739
Number of pages	4
ISBN (Print)	9781467364560
DOIs	https://doi.org/10.1109/ISBI.2013.6556580
Publication status	Published - 2013
Event	10th International Symposium on Biomedical Imaging: From Nano to Macro - Westin Hotel, San Francisco, CA., United States Duration: 7 Apr 2013 → 11 Apr 2013 http://www.biomedicalimaging.org/2013/

Conference

Conference	10th International Symposium on Biomedical Imaging: From Nano to Macro
Abbreviated title	ISBI 2013
Country/Territory	United States
City	San Francisco, CA.
Period	7/04/13 → 11/04/13
Internet address	http://www.biomedicalimaging.org/2013/

UN SDGs

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

Access to Document

10.1109/ISBI.2013.6556580

6 Citations
1 Paper

Classification of colorectal polyp regions in optical projection tomography
Li, W., Zhang, J., McKenna, S., Coats, M. & Carey, F., 9 Apr 2013.
Research output: Contribution to conference › Paper

4 Participation in conference

10th International Symposium on Biomedical Imaging: From Nano to Macro
Coats, M. (Participant)
9 Apr 2013
Activity: Participating in or organising an event types › Participation in conference
10th International Symposium on Biomedical Imaging: From Nano to Macro
McKenna, S. (Participant)
9 Apr 2013
Activity: Participating in or organising an event types › Participation in conference
10th International Symposium on Biomedical Imaging: From Nano to Macro
Carey, F. A. (Participant)
9 Apr 2013
Activity: Participating in or organising an event types › Participation in conference

Cite this

@inproceedings{918f1ae33c8a4462be192b1f2b3b56a4,

title = "Classification of colorectal polyp regions in optical projection tomography",

abstract = "The potential of optical projection tomography (OPT) to enhance colorectal polyp diagnosis is beginning to be explored. This paper presents, to the best of our knowledge, the first study on automatic image analysis of OPT images of colorectal polyps. 3D regions are classified using the bag of visual words framework and support vector machines. Independent subspace analysis is used to learn a domain-specific feature dictionary. This is compared to the use of raw patches (after random projection) and local binary patterns. Classification experiments (across patients) at the patch level and at the region level are presented using a set of 30 expert-annotated OPT images. Results show that accurate classification of 3D OPT image regions is feasible using this approach; regions of low-grade dysplasia and invasive cancer were discriminated with approximately 90% accuracy.",

author = "Wenqi Li and Jianguo Zhang and McKenna, {Stephen J.} and Maria Coats and Carey, {Frank A.}",

year = "2013",

doi = "10.1109/ISBI.2013.6556580",

language = "English",

isbn = "9781467364560 ",

pages = "736--739",

booktitle = "ISBI 2013",

publisher = "IEEE",

note = "10th International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2013 ; Conference date: 07-04-2013 Through 11-04-2013",

url = "http://www.biomedicalimaging.org/2013/",

}

Li, W, Zhang, J , McKenna, SJ, Coats, M & Carey, FA 2013, Classification of colorectal polyp regions in optical projection tomography. in ISBI 2013: IEEE 10th International Symposium on Biomedical Imaging: From Nano to Macro - Proceedings. IEEE, Piscataway, N.J., pp. 736-739, 10th International Symposium on Biomedical Imaging: From Nano to Macro, San Francisco, CA., United States, 7/04/13. https://doi.org/10.1109/ISBI.2013.6556580

TY - GEN

T1 - Classification of colorectal polyp regions in optical projection tomography

AU - Li, Wenqi

AU - Zhang, Jianguo

AU - McKenna, Stephen J.

AU - Coats, Maria

AU - Carey, Frank A.

PY - 2013

Y1 - 2013

N2 - The potential of optical projection tomography (OPT) to enhance colorectal polyp diagnosis is beginning to be explored. This paper presents, to the best of our knowledge, the first study on automatic image analysis of OPT images of colorectal polyps. 3D regions are classified using the bag of visual words framework and support vector machines. Independent subspace analysis is used to learn a domain-specific feature dictionary. This is compared to the use of raw patches (after random projection) and local binary patterns. Classification experiments (across patients) at the patch level and at the region level are presented using a set of 30 expert-annotated OPT images. Results show that accurate classification of 3D OPT image regions is feasible using this approach; regions of low-grade dysplasia and invasive cancer were discriminated with approximately 90% accuracy.

AB - The potential of optical projection tomography (OPT) to enhance colorectal polyp diagnosis is beginning to be explored. This paper presents, to the best of our knowledge, the first study on automatic image analysis of OPT images of colorectal polyps. 3D regions are classified using the bag of visual words framework and support vector machines. Independent subspace analysis is used to learn a domain-specific feature dictionary. This is compared to the use of raw patches (after random projection) and local binary patterns. Classification experiments (across patients) at the patch level and at the region level are presented using a set of 30 expert-annotated OPT images. Results show that accurate classification of 3D OPT image regions is feasible using this approach; regions of low-grade dysplasia and invasive cancer were discriminated with approximately 90% accuracy.

U2 - 10.1109/ISBI.2013.6556580

DO - 10.1109/ISBI.2013.6556580

M3 - Conference contribution

SN - 9781467364560

SP - 736

EP - 739

BT - ISBI 2013

PB - IEEE

CY - Piscataway, N.J.

T2 - 10th International Symposium on Biomedical Imaging: From Nano to Macro

Y2 - 7 April 2013 through 11 April 2013

ER -

Classification of colorectal polyp regions in optical projection tomography

Abstract

Conference

UN SDGs

Access to Document

Fingerprint

Research output

Classification of colorectal polyp regions in optical projection tomography

Activities

10th International Symposium on Biomedical Imaging: From Nano to Macro

10th International Symposium on Biomedical Imaging: From Nano to Macro

10th International Symposium on Biomedical Imaging: From Nano to Macro

Cite this