Abnormal image detection in endoscopy videos using a filter bank and local binary patterns

Ruwan Nawarathna, JungHwan Oh, Jayantha Muthukudage, Wallapak Tavanapong, Johnny Wong, Piet C. de Groen, Shou Jiang Tang

Research output: Contribution to journalArticle

44 Citations (Scopus)

Abstract

Finding mucosal abnormalities (e.g., erythema, blood, ulcer, erosion, and polyp) is one of the most essential tasks during endoscopy video review. Since these abnormalities typically appear in a small number of frames (around 5% of the total frame number), automated detection of frames with an abnormality can save physician[U+05F3]s time significantly. In this paper, we propose a new multi-texture analysis method that effectively discerns images showing mucosal abnormalities from the ones without any abnormality since most abnormalities in endoscopy images have textures that are clearly distinguishable from normal textures using an advanced image texture analysis method. The method uses a "texton histogram" of an image block as features. The histogram captures the distribution of different "textons" representing various textures in an endoscopy image. The textons are representative response vectors of an application of a combination of Leung and Malik (LM) filter bank (i.e., a set of image filters) and a set of Local Binary Patterns on the image. Our experimental results indicate that the proposed method achieves 92% recall and 91.8% specificity on wireless capsule endoscopy (WCE) images and 91% recall and 90.8% specificity on colonoscopy images.

Original languageEnglish (US)
Pages (from-to)70-91
Number of pages22
JournalNeurocomputing
Volume144
DOIs
StatePublished - Nov 20 2014

Fingerprint

Endoscopy
Filter banks
Image texture
Textures
Capsule Endoscopy
Colonoscopy
Erythema
Polyps
Ulcer
Erosion
Blood
Physicians

Keywords

  • Colonoscopy
  • Filter bank
  • Local binary pattern
  • Texton
  • Texton dictionary
  • Wireless capsule endoscopy

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Cognitive Neuroscience

Cite this

Nawarathna, R., Oh, J., Muthukudage, J., Tavanapong, W., Wong, J., de Groen, P. C., & Tang, S. J. (2014). Abnormal image detection in endoscopy videos using a filter bank and local binary patterns. Neurocomputing, 144, 70-91. https://doi.org/10.1016/j.neucom.2014.02.064

Abnormal image detection in endoscopy videos using a filter bank and local binary patterns. / Nawarathna, Ruwan; Oh, JungHwan; Muthukudage, Jayantha; Tavanapong, Wallapak; Wong, Johnny; de Groen, Piet C.; Tang, Shou Jiang.

In: Neurocomputing, Vol. 144, 20.11.2014, p. 70-91.

Research output: Contribution to journalArticle

Nawarathna, R, Oh, J, Muthukudage, J, Tavanapong, W, Wong, J, de Groen, PC & Tang, SJ 2014, 'Abnormal image detection in endoscopy videos using a filter bank and local binary patterns', Neurocomputing, vol. 144, pp. 70-91. https://doi.org/10.1016/j.neucom.2014.02.064
Nawarathna R, Oh J, Muthukudage J, Tavanapong W, Wong J, de Groen PC et al. Abnormal image detection in endoscopy videos using a filter bank and local binary patterns. Neurocomputing. 2014 Nov 20;144:70-91. https://doi.org/10.1016/j.neucom.2014.02.064
Nawarathna, Ruwan ; Oh, JungHwan ; Muthukudage, Jayantha ; Tavanapong, Wallapak ; Wong, Johnny ; de Groen, Piet C. ; Tang, Shou Jiang. / Abnormal image detection in endoscopy videos using a filter bank and local binary patterns. In: Neurocomputing. 2014 ; Vol. 144. pp. 70-91.
@article{4c4ba2c943db406aa1214e212bd7f064,
title = "Abnormal image detection in endoscopy videos using a filter bank and local binary patterns",
abstract = "Finding mucosal abnormalities (e.g., erythema, blood, ulcer, erosion, and polyp) is one of the most essential tasks during endoscopy video review. Since these abnormalities typically appear in a small number of frames (around 5{\%} of the total frame number), automated detection of frames with an abnormality can save physician[U+05F3]s time significantly. In this paper, we propose a new multi-texture analysis method that effectively discerns images showing mucosal abnormalities from the ones without any abnormality since most abnormalities in endoscopy images have textures that are clearly distinguishable from normal textures using an advanced image texture analysis method. The method uses a {"}texton histogram{"} of an image block as features. The histogram captures the distribution of different {"}textons{"} representing various textures in an endoscopy image. The textons are representative response vectors of an application of a combination of Leung and Malik (LM) filter bank (i.e., a set of image filters) and a set of Local Binary Patterns on the image. Our experimental results indicate that the proposed method achieves 92{\%} recall and 91.8{\%} specificity on wireless capsule endoscopy (WCE) images and 91{\%} recall and 90.8{\%} specificity on colonoscopy images.",
keywords = "Colonoscopy, Filter bank, Local binary pattern, Texton, Texton dictionary, Wireless capsule endoscopy",
author = "Ruwan Nawarathna and JungHwan Oh and Jayantha Muthukudage and Wallapak Tavanapong and Johnny Wong and {de Groen}, {Piet C.} and Tang, {Shou Jiang}",
year = "2014",
month = "11",
day = "20",
doi = "10.1016/j.neucom.2014.02.064",
language = "English (US)",
volume = "144",
pages = "70--91",
journal = "Neurocomputing",
issn = "0925-2312",
publisher = "Elsevier",

}

TY - JOUR

T1 - Abnormal image detection in endoscopy videos using a filter bank and local binary patterns

AU - Nawarathna, Ruwan

AU - Oh, JungHwan

AU - Muthukudage, Jayantha

AU - Tavanapong, Wallapak

AU - Wong, Johnny

AU - de Groen, Piet C.

AU - Tang, Shou Jiang

PY - 2014/11/20

Y1 - 2014/11/20

N2 - Finding mucosal abnormalities (e.g., erythema, blood, ulcer, erosion, and polyp) is one of the most essential tasks during endoscopy video review. Since these abnormalities typically appear in a small number of frames (around 5% of the total frame number), automated detection of frames with an abnormality can save physician[U+05F3]s time significantly. In this paper, we propose a new multi-texture analysis method that effectively discerns images showing mucosal abnormalities from the ones without any abnormality since most abnormalities in endoscopy images have textures that are clearly distinguishable from normal textures using an advanced image texture analysis method. The method uses a "texton histogram" of an image block as features. The histogram captures the distribution of different "textons" representing various textures in an endoscopy image. The textons are representative response vectors of an application of a combination of Leung and Malik (LM) filter bank (i.e., a set of image filters) and a set of Local Binary Patterns on the image. Our experimental results indicate that the proposed method achieves 92% recall and 91.8% specificity on wireless capsule endoscopy (WCE) images and 91% recall and 90.8% specificity on colonoscopy images.

AB - Finding mucosal abnormalities (e.g., erythema, blood, ulcer, erosion, and polyp) is one of the most essential tasks during endoscopy video review. Since these abnormalities typically appear in a small number of frames (around 5% of the total frame number), automated detection of frames with an abnormality can save physician[U+05F3]s time significantly. In this paper, we propose a new multi-texture analysis method that effectively discerns images showing mucosal abnormalities from the ones without any abnormality since most abnormalities in endoscopy images have textures that are clearly distinguishable from normal textures using an advanced image texture analysis method. The method uses a "texton histogram" of an image block as features. The histogram captures the distribution of different "textons" representing various textures in an endoscopy image. The textons are representative response vectors of an application of a combination of Leung and Malik (LM) filter bank (i.e., a set of image filters) and a set of Local Binary Patterns on the image. Our experimental results indicate that the proposed method achieves 92% recall and 91.8% specificity on wireless capsule endoscopy (WCE) images and 91% recall and 90.8% specificity on colonoscopy images.

KW - Colonoscopy

KW - Filter bank

KW - Local binary pattern

KW - Texton

KW - Texton dictionary

KW - Wireless capsule endoscopy

UR - http://www.scopus.com/inward/record.url?scp=84905975432&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84905975432&partnerID=8YFLogxK

U2 - 10.1016/j.neucom.2014.02.064

DO - 10.1016/j.neucom.2014.02.064

M3 - Article

AN - SCOPUS:84905975432

VL - 144

SP - 70

EP - 91

JO - Neurocomputing

JF - Neurocomputing

SN - 0925-2312

ER -