Texture analysis of optical coherence tomography images: Feasibility for tissue classification

Kirk W. Gossage, Tomasz S. Tkaczyk, Jeffrey J Rodriguez, Jennifer K Barton

Research output: Contribution to journalArticle

142 Citations (Scopus)

Abstract

Optical coherence tomography (OCT) acquires cross-sectional images of tissue by measuring back-reflected light. Images from in vivo OCT systems typically have a resolution of 10 to 15 mm, and are thus best suited for visualizing structures in the range of tens to hundreds of microns, such as tissue layers or glands. Many normal and abnormal tissues lack visible structures in this size range, so it may appear that OCT is unsuitable for identification of these tissues. However, examination of structure-poor OCT images reveals that they frequently display a characteristic texture that is due to speckle. We evaluated the application of statistical and spectral texture analysis techniques for differentiating tissue types based on the structural and speckle content in OCT images. Excellent correct classification rates were obtained when images had slight visual differences (mouse skin and fat, correct classification rates of 98.5 and 97.3%, respectively), and reasonable rates were obtained with nearly identical-appearing images (normal versus abnormal mouse lung, correct classification rates of 64.0 and 88.6%, respectively). This study shows that texture analysis of OCT images may be capable of differentiating tissue types Without reliance On Visible Structures.

Original languageEnglish (US)
Pages (from-to)570-575
Number of pages6
JournalJournal of Biomedical Optics
Volume8
Issue number3
DOIs
StatePublished - Jul 2003

Fingerprint

Optical tomography
Optical Coherence Tomography
textures
Textures
tomography
Tissue
Speckle
mice
glands
Oils and fats
fats
Skin
lungs
Fats
examination
Light
Lung

Keywords

  • Brodatz
  • Image processing
  • Lung
  • Speckle

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Radiological and Ultrasound Technology
  • Clinical Biochemistry
  • Atomic and Molecular Physics, and Optics

Cite this

Texture analysis of optical coherence tomography images : Feasibility for tissue classification. / Gossage, Kirk W.; Tkaczyk, Tomasz S.; Rodriguez, Jeffrey J; Barton, Jennifer K.

In: Journal of Biomedical Optics, Vol. 8, No. 3, 07.2003, p. 570-575.

Research output: Contribution to journalArticle

@article{61d273791c2a4c7183bf2e6ec2137750,
title = "Texture analysis of optical coherence tomography images: Feasibility for tissue classification",
abstract = "Optical coherence tomography (OCT) acquires cross-sectional images of tissue by measuring back-reflected light. Images from in vivo OCT systems typically have a resolution of 10 to 15 mm, and are thus best suited for visualizing structures in the range of tens to hundreds of microns, such as tissue layers or glands. Many normal and abnormal tissues lack visible structures in this size range, so it may appear that OCT is unsuitable for identification of these tissues. However, examination of structure-poor OCT images reveals that they frequently display a characteristic texture that is due to speckle. We evaluated the application of statistical and spectral texture analysis techniques for differentiating tissue types based on the structural and speckle content in OCT images. Excellent correct classification rates were obtained when images had slight visual differences (mouse skin and fat, correct classification rates of 98.5 and 97.3{\%}, respectively), and reasonable rates were obtained with nearly identical-appearing images (normal versus abnormal mouse lung, correct classification rates of 64.0 and 88.6{\%}, respectively). This study shows that texture analysis of OCT images may be capable of differentiating tissue types Without reliance On Visible Structures.",
keywords = "Brodatz, Image processing, Lung, Speckle",
author = "Gossage, {Kirk W.} and Tkaczyk, {Tomasz S.} and Rodriguez, {Jeffrey J} and Barton, {Jennifer K}",
year = "2003",
month = "7",
doi = "10.1117/1.1577575",
language = "English (US)",
volume = "8",
pages = "570--575",
journal = "Journal of Biomedical Optics",
issn = "1083-3668",
publisher = "SPIE",
number = "3",

}

TY - JOUR

T1 - Texture analysis of optical coherence tomography images

T2 - Feasibility for tissue classification

AU - Gossage, Kirk W.

AU - Tkaczyk, Tomasz S.

AU - Rodriguez, Jeffrey J

AU - Barton, Jennifer K

PY - 2003/7

Y1 - 2003/7

N2 - Optical coherence tomography (OCT) acquires cross-sectional images of tissue by measuring back-reflected light. Images from in vivo OCT systems typically have a resolution of 10 to 15 mm, and are thus best suited for visualizing structures in the range of tens to hundreds of microns, such as tissue layers or glands. Many normal and abnormal tissues lack visible structures in this size range, so it may appear that OCT is unsuitable for identification of these tissues. However, examination of structure-poor OCT images reveals that they frequently display a characteristic texture that is due to speckle. We evaluated the application of statistical and spectral texture analysis techniques for differentiating tissue types based on the structural and speckle content in OCT images. Excellent correct classification rates were obtained when images had slight visual differences (mouse skin and fat, correct classification rates of 98.5 and 97.3%, respectively), and reasonable rates were obtained with nearly identical-appearing images (normal versus abnormal mouse lung, correct classification rates of 64.0 and 88.6%, respectively). This study shows that texture analysis of OCT images may be capable of differentiating tissue types Without reliance On Visible Structures.

AB - Optical coherence tomography (OCT) acquires cross-sectional images of tissue by measuring back-reflected light. Images from in vivo OCT systems typically have a resolution of 10 to 15 mm, and are thus best suited for visualizing structures in the range of tens to hundreds of microns, such as tissue layers or glands. Many normal and abnormal tissues lack visible structures in this size range, so it may appear that OCT is unsuitable for identification of these tissues. However, examination of structure-poor OCT images reveals that they frequently display a characteristic texture that is due to speckle. We evaluated the application of statistical and spectral texture analysis techniques for differentiating tissue types based on the structural and speckle content in OCT images. Excellent correct classification rates were obtained when images had slight visual differences (mouse skin and fat, correct classification rates of 98.5 and 97.3%, respectively), and reasonable rates were obtained with nearly identical-appearing images (normal versus abnormal mouse lung, correct classification rates of 64.0 and 88.6%, respectively). This study shows that texture analysis of OCT images may be capable of differentiating tissue types Without reliance On Visible Structures.

KW - Brodatz

KW - Image processing

KW - Lung

KW - Speckle

UR - http://www.scopus.com/inward/record.url?scp=0141533017&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0141533017&partnerID=8YFLogxK

U2 - 10.1117/1.1577575

DO - 10.1117/1.1577575

M3 - Article

C2 - 12880366

AN - SCOPUS:0141533017

VL - 8

SP - 570

EP - 575

JO - Journal of Biomedical Optics

JF - Journal of Biomedical Optics

SN - 1083-3668

IS - 3

ER -