Automated image alignment and segmentation to follow progression of geographic atrophy in age-related macular degeneration

David J. Ramsey, Janet S. Sunness, Poorva Malviya, Carol Applegate, Gregory Hager, James Handa

Research output: Contribution to journalArticle

Abstract

PURPOSE: To develop a computer-based image segmentation method for standardizing the quantification of geographic atrophy (GA). METHODS: The authors present an automated image segmentation method based on the fuzzy c-means clustering algorithm for the detection of GA lesions. The method is evaluated by comparing computerized segmentation against outlines of GA drawn by an expert grader for a longitudinal series of fundus autofluorescence images with paired 30° color fundus photographs for 10 patients. RESULTS: The automated segmentation method showed excellent agreement with an expert grader for fundus autofluorescence images, achieving a performance level of 94 ± 5% sensitivity and 98 ± 2% specificity on a per-pixel basis for the detection of GA area, but performed less well on color fundus photographs with a sensitivity of 47 ± 26% and specificity of 98 ± 2%. The segmentation algorithm identified 75 ± 16% of the GA border correctly in fundus autofluorescence images compared with just 42 ± 25% for color fundus photographs. CONCLUSION: The results of this study demonstrate a promising computerized segmentation method that may enhance the reproducibility of GA measurement and provide an objective strategy to assist an expert in the grading of images.

Original languageEnglish (US)
Pages (from-to)1296-1307
Number of pages12
JournalRetina
Volume34
Issue number7
DOIs
StatePublished - 2014

Fingerprint

Geographic Atrophy
Macular Degeneration
Color
Cluster Analysis

Keywords

  • age-related macular degeneration
  • automated image segmentation
  • fundus autofluorescence
  • geographic atrophy

ASJC Scopus subject areas

  • Ophthalmology
  • Medicine(all)

Cite this

Automated image alignment and segmentation to follow progression of geographic atrophy in age-related macular degeneration. / Ramsey, David J.; Sunness, Janet S.; Malviya, Poorva; Applegate, Carol; Hager, Gregory; Handa, James.

In: Retina, Vol. 34, No. 7, 2014, p. 1296-1307.

Research output: Contribution to journalArticle

Ramsey, David J. ; Sunness, Janet S. ; Malviya, Poorva ; Applegate, Carol ; Hager, Gregory ; Handa, James. / Automated image alignment and segmentation to follow progression of geographic atrophy in age-related macular degeneration. In: Retina. 2014 ; Vol. 34, No. 7. pp. 1296-1307.
@article{4acbe9462a2d4a5ba7219653c7609bd9,
title = "Automated image alignment and segmentation to follow progression of geographic atrophy in age-related macular degeneration",
abstract = "PURPOSE: To develop a computer-based image segmentation method for standardizing the quantification of geographic atrophy (GA). METHODS: The authors present an automated image segmentation method based on the fuzzy c-means clustering algorithm for the detection of GA lesions. The method is evaluated by comparing computerized segmentation against outlines of GA drawn by an expert grader for a longitudinal series of fundus autofluorescence images with paired 30° color fundus photographs for 10 patients. RESULTS: The automated segmentation method showed excellent agreement with an expert grader for fundus autofluorescence images, achieving a performance level of 94 ± 5{\%} sensitivity and 98 ± 2{\%} specificity on a per-pixel basis for the detection of GA area, but performed less well on color fundus photographs with a sensitivity of 47 ± 26{\%} and specificity of 98 ± 2{\%}. The segmentation algorithm identified 75 ± 16{\%} of the GA border correctly in fundus autofluorescence images compared with just 42 ± 25{\%} for color fundus photographs. CONCLUSION: The results of this study demonstrate a promising computerized segmentation method that may enhance the reproducibility of GA measurement and provide an objective strategy to assist an expert in the grading of images.",
keywords = "age-related macular degeneration, automated image segmentation, fundus autofluorescence, geographic atrophy",
author = "Ramsey, {David J.} and Sunness, {Janet S.} and Poorva Malviya and Carol Applegate and Gregory Hager and James Handa",
year = "2014",
doi = "10.1097/IAE.0000000000000069",
language = "English (US)",
volume = "34",
pages = "1296--1307",
journal = "Retina",
issn = "0275-004X",
publisher = "Lippincott Williams and Wilkins",
number = "7",

}

TY - JOUR

T1 - Automated image alignment and segmentation to follow progression of geographic atrophy in age-related macular degeneration

AU - Ramsey, David J.

AU - Sunness, Janet S.

AU - Malviya, Poorva

AU - Applegate, Carol

AU - Hager, Gregory

AU - Handa, James

PY - 2014

Y1 - 2014

N2 - PURPOSE: To develop a computer-based image segmentation method for standardizing the quantification of geographic atrophy (GA). METHODS: The authors present an automated image segmentation method based on the fuzzy c-means clustering algorithm for the detection of GA lesions. The method is evaluated by comparing computerized segmentation against outlines of GA drawn by an expert grader for a longitudinal series of fundus autofluorescence images with paired 30° color fundus photographs for 10 patients. RESULTS: The automated segmentation method showed excellent agreement with an expert grader for fundus autofluorescence images, achieving a performance level of 94 ± 5% sensitivity and 98 ± 2% specificity on a per-pixel basis for the detection of GA area, but performed less well on color fundus photographs with a sensitivity of 47 ± 26% and specificity of 98 ± 2%. The segmentation algorithm identified 75 ± 16% of the GA border correctly in fundus autofluorescence images compared with just 42 ± 25% for color fundus photographs. CONCLUSION: The results of this study demonstrate a promising computerized segmentation method that may enhance the reproducibility of GA measurement and provide an objective strategy to assist an expert in the grading of images.

AB - PURPOSE: To develop a computer-based image segmentation method for standardizing the quantification of geographic atrophy (GA). METHODS: The authors present an automated image segmentation method based on the fuzzy c-means clustering algorithm for the detection of GA lesions. The method is evaluated by comparing computerized segmentation against outlines of GA drawn by an expert grader for a longitudinal series of fundus autofluorescence images with paired 30° color fundus photographs for 10 patients. RESULTS: The automated segmentation method showed excellent agreement with an expert grader for fundus autofluorescence images, achieving a performance level of 94 ± 5% sensitivity and 98 ± 2% specificity on a per-pixel basis for the detection of GA area, but performed less well on color fundus photographs with a sensitivity of 47 ± 26% and specificity of 98 ± 2%. The segmentation algorithm identified 75 ± 16% of the GA border correctly in fundus autofluorescence images compared with just 42 ± 25% for color fundus photographs. CONCLUSION: The results of this study demonstrate a promising computerized segmentation method that may enhance the reproducibility of GA measurement and provide an objective strategy to assist an expert in the grading of images.

KW - age-related macular degeneration

KW - automated image segmentation

KW - fundus autofluorescence

KW - geographic atrophy

UR - http://www.scopus.com/inward/record.url?scp=84903270918&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84903270918&partnerID=8YFLogxK

U2 - 10.1097/IAE.0000000000000069

DO - 10.1097/IAE.0000000000000069

M3 - Article

C2 - 24398699

AN - SCOPUS:84903270918

VL - 34

SP - 1296

EP - 1307

JO - Retina

JF - Retina

SN - 0275-004X

IS - 7

ER -