Automated detection of microaneurysms in digital red-free photographs: a diabetic retinopathy screening tool

J H Hipwell, F Strachan, J A Olson, K C McHardy, P F Sharp, J V Forrester

Research output: Contribution to journalArticle

122 Citations (Scopus)

Abstract

Aims To develop a technique to detect microaneurysms automatically in 50 degrees digital red-free fundus photographs and evaluate its performance as a tool for screening diabetic patients for retinopathy,

Methods Candidate microaneurysms are extracted, after the image has been modified to remove variations In background intensity, by algorithms that enhance small round features. Each microaneurysm candidate is then classified according to its intensity and size by the application of a sec of rules derived from a training set of 102 images.

Results When 3783 individual images were analysed and the results compared with the opinion of a clinical research fellow examining the same images, the program achieved a sensitivity of 81% and a specificity of 93% for the detection of images containing microaneurysms, Nine hundred and twenty-five sets of 4 images per patient were then analysed and the total number of microaneurysms detected compared with the overall patient retinopathy grade derived by the clinician examining the same images. In this context, intended to mimic a screening situation, the program achieved a sensitivity of 85% and a specificity of 76% for the detection of patients with (any) retinopathy (positive predictive value 0.71, negative predictive value 0,88),

Conclusions An automated technique was developed to detect retinopathy in digital red-free fundus images that can form part of a diabetic retinopathy screening programme. It is believed that it can perform a useful role in this context identifying images worthy of closer inspection or eliminating 50% or more of the screening population who have no retinopathy.

Original languageEnglish
Pages (from-to)588-594
Number of pages7
JournalDiabetic Medicine
Volume17
Publication statusPublished - 2000

Keywords

  • digital retinal photography
  • microaneurysms
  • image processing
  • EURODIAB IDDM COMPLICATIONS
  • FLUORESCEIN ANGIOGRAMS
  • FUNDUS
  • QUANTIFICATION
  • PROGRESSION
  • SEVERITY
  • SYSTEM

Cite this

Hipwell, J. H., Strachan, F., Olson, J. A., McHardy, K. C., Sharp, P. F., & Forrester, J. V. (2000). Automated detection of microaneurysms in digital red-free photographs: a diabetic retinopathy screening tool. Diabetic Medicine, 17, 588-594.

Automated detection of microaneurysms in digital red-free photographs: a diabetic retinopathy screening tool. / Hipwell, J H ; Strachan, F ; Olson, J A ; McHardy, K C ; Sharp, P F ; Forrester, J V .

In: Diabetic Medicine, Vol. 17, 2000, p. 588-594.

Research output: Contribution to journalArticle

Hipwell, JH, Strachan, F, Olson, JA, McHardy, KC, Sharp, PF & Forrester, JV 2000, 'Automated detection of microaneurysms in digital red-free photographs: a diabetic retinopathy screening tool' Diabetic Medicine, vol. 17, pp. 588-594.
Hipwell JH, Strachan F, Olson JA, McHardy KC, Sharp PF, Forrester JV. Automated detection of microaneurysms in digital red-free photographs: a diabetic retinopathy screening tool. Diabetic Medicine. 2000;17:588-594.
Hipwell, J H ; Strachan, F ; Olson, J A ; McHardy, K C ; Sharp, P F ; Forrester, J V . / Automated detection of microaneurysms in digital red-free photographs: a diabetic retinopathy screening tool. In: Diabetic Medicine. 2000 ; Vol. 17. pp. 588-594.
@article{54d285ccc05b4543aeeec4e736c60210,
title = "Automated detection of microaneurysms in digital red-free photographs: a diabetic retinopathy screening tool",
abstract = "Aims To develop a technique to detect microaneurysms automatically in 50 degrees digital red-free fundus photographs and evaluate its performance as a tool for screening diabetic patients for retinopathy,Methods Candidate microaneurysms are extracted, after the image has been modified to remove variations In background intensity, by algorithms that enhance small round features. Each microaneurysm candidate is then classified according to its intensity and size by the application of a sec of rules derived from a training set of 102 images.Results When 3783 individual images were analysed and the results compared with the opinion of a clinical research fellow examining the same images, the program achieved a sensitivity of 81{\%} and a specificity of 93{\%} for the detection of images containing microaneurysms, Nine hundred and twenty-five sets of 4 images per patient were then analysed and the total number of microaneurysms detected compared with the overall patient retinopathy grade derived by the clinician examining the same images. In this context, intended to mimic a screening situation, the program achieved a sensitivity of 85{\%} and a specificity of 76{\%} for the detection of patients with (any) retinopathy (positive predictive value 0.71, negative predictive value 0,88),Conclusions An automated technique was developed to detect retinopathy in digital red-free fundus images that can form part of a diabetic retinopathy screening programme. It is believed that it can perform a useful role in this context identifying images worthy of closer inspection or eliminating 50{\%} or more of the screening population who have no retinopathy.",
keywords = "digital retinal photography, microaneurysms, image processing, EURODIAB IDDM COMPLICATIONS, FLUORESCEIN ANGIOGRAMS, FUNDUS, QUANTIFICATION, PROGRESSION, SEVERITY, SYSTEM",
author = "Hipwell, {J H} and F Strachan and Olson, {J A} and McHardy, {K C} and Sharp, {P F} and Forrester, {J V}",
year = "2000",
language = "English",
volume = "17",
pages = "588--594",
journal = "Diabetic Medicine",
issn = "0742-3071",
publisher = "Wiley",

}

TY - JOUR

T1 - Automated detection of microaneurysms in digital red-free photographs: a diabetic retinopathy screening tool

AU - Hipwell, J H

AU - Strachan, F

AU - Olson, J A

AU - McHardy, K C

AU - Sharp, P F

AU - Forrester, J V

PY - 2000

Y1 - 2000

N2 - Aims To develop a technique to detect microaneurysms automatically in 50 degrees digital red-free fundus photographs and evaluate its performance as a tool for screening diabetic patients for retinopathy,Methods Candidate microaneurysms are extracted, after the image has been modified to remove variations In background intensity, by algorithms that enhance small round features. Each microaneurysm candidate is then classified according to its intensity and size by the application of a sec of rules derived from a training set of 102 images.Results When 3783 individual images were analysed and the results compared with the opinion of a clinical research fellow examining the same images, the program achieved a sensitivity of 81% and a specificity of 93% for the detection of images containing microaneurysms, Nine hundred and twenty-five sets of 4 images per patient were then analysed and the total number of microaneurysms detected compared with the overall patient retinopathy grade derived by the clinician examining the same images. In this context, intended to mimic a screening situation, the program achieved a sensitivity of 85% and a specificity of 76% for the detection of patients with (any) retinopathy (positive predictive value 0.71, negative predictive value 0,88),Conclusions An automated technique was developed to detect retinopathy in digital red-free fundus images that can form part of a diabetic retinopathy screening programme. It is believed that it can perform a useful role in this context identifying images worthy of closer inspection or eliminating 50% or more of the screening population who have no retinopathy.

AB - Aims To develop a technique to detect microaneurysms automatically in 50 degrees digital red-free fundus photographs and evaluate its performance as a tool for screening diabetic patients for retinopathy,Methods Candidate microaneurysms are extracted, after the image has been modified to remove variations In background intensity, by algorithms that enhance small round features. Each microaneurysm candidate is then classified according to its intensity and size by the application of a sec of rules derived from a training set of 102 images.Results When 3783 individual images were analysed and the results compared with the opinion of a clinical research fellow examining the same images, the program achieved a sensitivity of 81% and a specificity of 93% for the detection of images containing microaneurysms, Nine hundred and twenty-five sets of 4 images per patient were then analysed and the total number of microaneurysms detected compared with the overall patient retinopathy grade derived by the clinician examining the same images. In this context, intended to mimic a screening situation, the program achieved a sensitivity of 85% and a specificity of 76% for the detection of patients with (any) retinopathy (positive predictive value 0.71, negative predictive value 0,88),Conclusions An automated technique was developed to detect retinopathy in digital red-free fundus images that can form part of a diabetic retinopathy screening programme. It is believed that it can perform a useful role in this context identifying images worthy of closer inspection or eliminating 50% or more of the screening population who have no retinopathy.

KW - digital retinal photography

KW - microaneurysms

KW - image processing

KW - EURODIAB IDDM COMPLICATIONS

KW - FLUORESCEIN ANGIOGRAMS

KW - FUNDUS

KW - QUANTIFICATION

KW - PROGRESSION

KW - SEVERITY

KW - SYSTEM

M3 - Article

VL - 17

SP - 588

EP - 594

JO - Diabetic Medicine

JF - Diabetic Medicine

SN - 0742-3071

ER -