Automated segmentation of optical coherence tomography images

C. Kharmyssov, Wai Ko, J. R. Kim

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

We propose a fast and accurate automated algorithm to segment retinal pigment epithelium and internal limiting membrane layers from spectral domain optical coherence tomography (SDOCT) B-scan images. A hybrid algorithm, which combines intensity thresholding and graph-based algorithms, was used to process and analyze SDOCT radial scans (120 B scans) images obtained from twenty patients. The relative difference in position of the layers segmented by the proposed hybrid algorithm and by the clinical expert was 1.49% ± 0.01%. The processing time of the hybrid algorithm was 9.3 s for six B scans. Dice's coefficient of the hybrid algorithm was 96.7% ± 1.6%. The proposed hybrid algorithm for the segmentation of SDOCT images had good agreement with manual segmentation and reduced processing time.

Original languageEnglish
Article number011701
JournalChinese Optics Letters
Volume17
Issue number1
DOIs
Publication statusPublished - Jan 10 2019

Fingerprint

Optical tomography
tomography
Retinal Pigments
epithelium
Processing
pigments
Pigments
membranes
Membranes
coefficients

ASJC Scopus subject areas

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

Cite this

Automated segmentation of optical coherence tomography images. / Kharmyssov, C.; Ko, Wai; Kim, J. R.

In: Chinese Optics Letters, Vol. 17, No. 1, 011701, 10.01.2019.

Research output: Contribution to journalArticle

Kharmyssov, C. ; Ko, Wai ; Kim, J. R. / Automated segmentation of optical coherence tomography images. In: Chinese Optics Letters. 2019 ; Vol. 17, No. 1.
@article{9ca79330e1df4d0e9d878d809bcdcb8f,
title = "Automated segmentation of optical coherence tomography images",
abstract = "We propose a fast and accurate automated algorithm to segment retinal pigment epithelium and internal limiting membrane layers from spectral domain optical coherence tomography (SDOCT) B-scan images. A hybrid algorithm, which combines intensity thresholding and graph-based algorithms, was used to process and analyze SDOCT radial scans (120 B scans) images obtained from twenty patients. The relative difference in position of the layers segmented by the proposed hybrid algorithm and by the clinical expert was 1.49{\%} ± 0.01{\%}. The processing time of the hybrid algorithm was 9.3 s for six B scans. Dice's coefficient of the hybrid algorithm was 96.7{\%} ± 1.6{\%}. The proposed hybrid algorithm for the segmentation of SDOCT images had good agreement with manual segmentation and reduced processing time.",
author = "C. Kharmyssov and Wai Ko and Kim, {J. R.}",
year = "2019",
month = "1",
day = "10",
doi = "10.3788/COL201917.011701",
language = "English",
volume = "17",
journal = "Chinese Optics Letters",
issn = "1671-7694",
publisher = "Science Press",
number = "1",

}

TY - JOUR

T1 - Automated segmentation of optical coherence tomography images

AU - Kharmyssov, C.

AU - Ko, Wai

AU - Kim, J. R.

PY - 2019/1/10

Y1 - 2019/1/10

N2 - We propose a fast and accurate automated algorithm to segment retinal pigment epithelium and internal limiting membrane layers from spectral domain optical coherence tomography (SDOCT) B-scan images. A hybrid algorithm, which combines intensity thresholding and graph-based algorithms, was used to process and analyze SDOCT radial scans (120 B scans) images obtained from twenty patients. The relative difference in position of the layers segmented by the proposed hybrid algorithm and by the clinical expert was 1.49% ± 0.01%. The processing time of the hybrid algorithm was 9.3 s for six B scans. Dice's coefficient of the hybrid algorithm was 96.7% ± 1.6%. The proposed hybrid algorithm for the segmentation of SDOCT images had good agreement with manual segmentation and reduced processing time.

AB - We propose a fast and accurate automated algorithm to segment retinal pigment epithelium and internal limiting membrane layers from spectral domain optical coherence tomography (SDOCT) B-scan images. A hybrid algorithm, which combines intensity thresholding and graph-based algorithms, was used to process and analyze SDOCT radial scans (120 B scans) images obtained from twenty patients. The relative difference in position of the layers segmented by the proposed hybrid algorithm and by the clinical expert was 1.49% ± 0.01%. The processing time of the hybrid algorithm was 9.3 s for six B scans. Dice's coefficient of the hybrid algorithm was 96.7% ± 1.6%. The proposed hybrid algorithm for the segmentation of SDOCT images had good agreement with manual segmentation and reduced processing time.

UR - http://www.scopus.com/inward/record.url?scp=85063052416&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85063052416&partnerID=8YFLogxK

U2 - 10.3788/COL201917.011701

DO - 10.3788/COL201917.011701

M3 - Article

VL - 17

JO - Chinese Optics Letters

JF - Chinese Optics Letters

SN - 1671-7694

IS - 1

M1 - 011701

ER -