Improving image segmentation performance and quantitative analysis via a computer-aided grading methodology for optical coherence tomography retinal image analysis

Delia Cabrera Debuc, Harry M. Salinas, Sudarshan Ranganathan, Erika Tátrai, Wei Gao, Meixiao Shen, Jianhua Wang, G. M. Somfai, Carmen A. Puliafito

Research output: Contribution to journalArticle

20 Citations (Scopus)

Abstract

We demonstrate quantitative analysis and error correction of optical coherence tomography (OCT) retinal images by using a custom-built, computer-aided grading methodology. A total of 60 Stratus OCT (Carl Zeiss Meditec, Dublin, California) B-scans collected from ten normal healthy eyes are analyzed by two independent graders. The average retinal thickness per macular region is compared with the automated Stratus OCT results. Intergrader and intragrader reproducibility is calculated by Bland-Altman plots of the mean difference between both gradings and by Pearson correlation coefficients. In addition, the correlation between Stratus OCT and our methodology-derived thickness is also presented. The mean thickness difference between Stratus OCT and our methodology is 6.53 μm and 26.71 μm when using the inner segment/outer segment (IS/OS) junction and outer segment/retinal pigment epithelium (OS/RPE) junction as the outer retinal border, respectively. Overall, the median of the thickness differences as a percentage of the mean thickness is less than 1% and 2% for the intragrader and intergrader reproducibility test, respectively. The measurement accuracy range of the OCT retinal image analysis (OCTRIMA) algorithm is between 0.27 and 1.47 μm and 0.6 and 1.76 μm for the intragrader and intergrader reproducibility tests, respectively. Pearson correlation coefficients demonstrate R 2>0.98 for all Early Treatment Diabetic Retinopathy Study (ETDRS) regions. Our methodology facilitates a more robust and localized quantification of the retinal structure in normal healthy controls and patients with clinically significant intraretinal features.

Original languageEnglish
Article number046015
JournalJournal of Biomedical Optics
Volume15
Issue number4
DOIs
Publication statusPublished - Jul 2010

Fingerprint

retinal images
Optical tomography
image analysis
Image segmentation
Image analysis
quantitative analysis
tomography
methodology
Chemical analysis
correlation coefficients
Retinal Pigments
epithelium
Error correction
pigments
borders
Pigments
plots

Keywords

  • Image processing
  • Optical coherence tomography
  • Retinal image analysis
  • Segmentation

ASJC Scopus subject areas

  • Biomedical Engineering
  • Biomaterials
  • Electronic, Optical and Magnetic Materials
  • Atomic and Molecular Physics, and Optics

Cite this

Improving image segmentation performance and quantitative analysis via a computer-aided grading methodology for optical coherence tomography retinal image analysis. / Debuc, Delia Cabrera; Salinas, Harry M.; Ranganathan, Sudarshan; Tátrai, Erika; Gao, Wei; Shen, Meixiao; Wang, Jianhua; Somfai, G. M.; Puliafito, Carmen A.

In: Journal of Biomedical Optics, Vol. 15, No. 4, 046015, 07.2010.

Research output: Contribution to journalArticle

Debuc, Delia Cabrera ; Salinas, Harry M. ; Ranganathan, Sudarshan ; Tátrai, Erika ; Gao, Wei ; Shen, Meixiao ; Wang, Jianhua ; Somfai, G. M. ; Puliafito, Carmen A. / Improving image segmentation performance and quantitative analysis via a computer-aided grading methodology for optical coherence tomography retinal image analysis. In: Journal of Biomedical Optics. 2010 ; Vol. 15, No. 4.
@article{07b184dd4b464ff393f39042de245810,
title = "Improving image segmentation performance and quantitative analysis via a computer-aided grading methodology for optical coherence tomography retinal image analysis",
abstract = "We demonstrate quantitative analysis and error correction of optical coherence tomography (OCT) retinal images by using a custom-built, computer-aided grading methodology. A total of 60 Stratus OCT (Carl Zeiss Meditec, Dublin, California) B-scans collected from ten normal healthy eyes are analyzed by two independent graders. The average retinal thickness per macular region is compared with the automated Stratus OCT results. Intergrader and intragrader reproducibility is calculated by Bland-Altman plots of the mean difference between both gradings and by Pearson correlation coefficients. In addition, the correlation between Stratus OCT and our methodology-derived thickness is also presented. The mean thickness difference between Stratus OCT and our methodology is 6.53 μm and 26.71 μm when using the inner segment/outer segment (IS/OS) junction and outer segment/retinal pigment epithelium (OS/RPE) junction as the outer retinal border, respectively. Overall, the median of the thickness differences as a percentage of the mean thickness is less than 1{\%} and 2{\%} for the intragrader and intergrader reproducibility test, respectively. The measurement accuracy range of the OCT retinal image analysis (OCTRIMA) algorithm is between 0.27 and 1.47 μm and 0.6 and 1.76 μm for the intragrader and intergrader reproducibility tests, respectively. Pearson correlation coefficients demonstrate R 2>0.98 for all Early Treatment Diabetic Retinopathy Study (ETDRS) regions. Our methodology facilitates a more robust and localized quantification of the retinal structure in normal healthy controls and patients with clinically significant intraretinal features.",
keywords = "Image processing, Optical coherence tomography, Retinal image analysis, Segmentation",
author = "Debuc, {Delia Cabrera} and Salinas, {Harry M.} and Sudarshan Ranganathan and Erika T{\'a}trai and Wei Gao and Meixiao Shen and Jianhua Wang and Somfai, {G. M.} and Puliafito, {Carmen A.}",
year = "2010",
month = "7",
doi = "10.1117/1.3470116",
language = "English",
volume = "15",
journal = "Journal of Biomedical Optics",
issn = "1083-3668",
publisher = "SPIE",
number = "4",

}

TY - JOUR

T1 - Improving image segmentation performance and quantitative analysis via a computer-aided grading methodology for optical coherence tomography retinal image analysis

AU - Debuc, Delia Cabrera

AU - Salinas, Harry M.

AU - Ranganathan, Sudarshan

AU - Tátrai, Erika

AU - Gao, Wei

AU - Shen, Meixiao

AU - Wang, Jianhua

AU - Somfai, G. M.

AU - Puliafito, Carmen A.

PY - 2010/7

Y1 - 2010/7

N2 - We demonstrate quantitative analysis and error correction of optical coherence tomography (OCT) retinal images by using a custom-built, computer-aided grading methodology. A total of 60 Stratus OCT (Carl Zeiss Meditec, Dublin, California) B-scans collected from ten normal healthy eyes are analyzed by two independent graders. The average retinal thickness per macular region is compared with the automated Stratus OCT results. Intergrader and intragrader reproducibility is calculated by Bland-Altman plots of the mean difference between both gradings and by Pearson correlation coefficients. In addition, the correlation between Stratus OCT and our methodology-derived thickness is also presented. The mean thickness difference between Stratus OCT and our methodology is 6.53 μm and 26.71 μm when using the inner segment/outer segment (IS/OS) junction and outer segment/retinal pigment epithelium (OS/RPE) junction as the outer retinal border, respectively. Overall, the median of the thickness differences as a percentage of the mean thickness is less than 1% and 2% for the intragrader and intergrader reproducibility test, respectively. The measurement accuracy range of the OCT retinal image analysis (OCTRIMA) algorithm is between 0.27 and 1.47 μm and 0.6 and 1.76 μm for the intragrader and intergrader reproducibility tests, respectively. Pearson correlation coefficients demonstrate R 2>0.98 for all Early Treatment Diabetic Retinopathy Study (ETDRS) regions. Our methodology facilitates a more robust and localized quantification of the retinal structure in normal healthy controls and patients with clinically significant intraretinal features.

AB - We demonstrate quantitative analysis and error correction of optical coherence tomography (OCT) retinal images by using a custom-built, computer-aided grading methodology. A total of 60 Stratus OCT (Carl Zeiss Meditec, Dublin, California) B-scans collected from ten normal healthy eyes are analyzed by two independent graders. The average retinal thickness per macular region is compared with the automated Stratus OCT results. Intergrader and intragrader reproducibility is calculated by Bland-Altman plots of the mean difference between both gradings and by Pearson correlation coefficients. In addition, the correlation between Stratus OCT and our methodology-derived thickness is also presented. The mean thickness difference between Stratus OCT and our methodology is 6.53 μm and 26.71 μm when using the inner segment/outer segment (IS/OS) junction and outer segment/retinal pigment epithelium (OS/RPE) junction as the outer retinal border, respectively. Overall, the median of the thickness differences as a percentage of the mean thickness is less than 1% and 2% for the intragrader and intergrader reproducibility test, respectively. The measurement accuracy range of the OCT retinal image analysis (OCTRIMA) algorithm is between 0.27 and 1.47 μm and 0.6 and 1.76 μm for the intragrader and intergrader reproducibility tests, respectively. Pearson correlation coefficients demonstrate R 2>0.98 for all Early Treatment Diabetic Retinopathy Study (ETDRS) regions. Our methodology facilitates a more robust and localized quantification of the retinal structure in normal healthy controls and patients with clinically significant intraretinal features.

KW - Image processing

KW - Optical coherence tomography

KW - Retinal image analysis

KW - Segmentation

UR - http://www.scopus.com/inward/record.url?scp=79952198809&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79952198809&partnerID=8YFLogxK

U2 - 10.1117/1.3470116

DO - 10.1117/1.3470116

M3 - Article

C2 - 20799817

AN - SCOPUS:79952198809

VL - 15

JO - Journal of Biomedical Optics

JF - Journal of Biomedical Optics

SN - 1083-3668

IS - 4

M1 - 046015

ER -