A fully automated comparative microaneurysm digital detection system

Michael J Cree, John A Olson, Kenneth C McHardy, Peter F Sharp, John V Forrester

Research output: Contribution to journalArticle

127 Citations (Scopus)

Abstract

A fully automated digital image processing system, which provides an objective and repeatable way to quantify microaneurysms in digitised fluorescein angiograms, has been developed. The automated computer processing includes registration of same-eye retinal images for serial studies, cutting of regions-of-interest centred on the fovea, the detection of microaneurysms and the comparison of serial images for microaneurysm turnover. The microaneurysm detector was trained against a database of 68 images of patients with diabetes containing 394 true microaneurysms, as identified by an ophthalmologist. The microaneurysm detector achieved 82% sensitivity with 2.0 false-positives per image. An independent test set, comprising 20 images containing 297 true microaneurysms, was used to compare the microaneurysm detector with clinicians. The microaneurysm detector achieved a sensitivity of 82% for 5.7 false-positives per image, whereas the clinician receiver-operator-characteristic (ROC) curve gives 3.2 false-positives per image at a sensitivity of 82%. It is concluded that the computer system can reliably detect microaneurysms. The advantages of the computer system include objectivity, repeatability, speed and full automation.
Original languageEnglish
Pages (from-to)622-628
Number of pages7
JournalEye
Volume11
Issue number5
DOIs
Publication statusPublished - Sep 1997

Fingerprint

Computer Systems
Microaneurysm
Automation
Fluorescein
Angiography
Databases
Ophthalmologists

Keywords

  • Aneurysm
  • Diabetic Retinopathy
  • Disease Progression
  • False Positive Reactions
  • Fluorescein Angiography
  • Follow-Up Studies
  • Humans
  • Image Interpretation, Computer-Assisted
  • ROC Curve
  • Sensitivity and Specificity
  • Microaneurysm
  • Diabetes
  • Retinopathy
  • Computer
  • Digitisation

Cite this

A fully automated comparative microaneurysm digital detection system. / Cree, Michael J; Olson, John A; McHardy, Kenneth C; Sharp, Peter F; Forrester, John V.

In: Eye, Vol. 11, No. 5, 09.1997, p. 622-628.

Research output: Contribution to journalArticle

Cree, Michael J ; Olson, John A ; McHardy, Kenneth C ; Sharp, Peter F ; Forrester, John V. / A fully automated comparative microaneurysm digital detection system. In: Eye. 1997 ; Vol. 11, No. 5. pp. 622-628.
@article{de2271f145ec44be9e93efd423afbb04,
title = "A fully automated comparative microaneurysm digital detection system",
abstract = "A fully automated digital image processing system, which provides an objective and repeatable way to quantify microaneurysms in digitised fluorescein angiograms, has been developed. The automated computer processing includes registration of same-eye retinal images for serial studies, cutting of regions-of-interest centred on the fovea, the detection of microaneurysms and the comparison of serial images for microaneurysm turnover. The microaneurysm detector was trained against a database of 68 images of patients with diabetes containing 394 true microaneurysms, as identified by an ophthalmologist. The microaneurysm detector achieved 82{\%} sensitivity with 2.0 false-positives per image. An independent test set, comprising 20 images containing 297 true microaneurysms, was used to compare the microaneurysm detector with clinicians. The microaneurysm detector achieved a sensitivity of 82{\%} for 5.7 false-positives per image, whereas the clinician receiver-operator-characteristic (ROC) curve gives 3.2 false-positives per image at a sensitivity of 82{\%}. It is concluded that the computer system can reliably detect microaneurysms. The advantages of the computer system include objectivity, repeatability, speed and full automation.",
keywords = "Aneurysm, Diabetic Retinopathy, Disease Progression, False Positive Reactions, Fluorescein Angiography, Follow-Up Studies, Humans, Image Interpretation, Computer-Assisted, ROC Curve, Sensitivity and Specificity, Microaneurysm, Diabetes, Retinopathy, Computer, Digitisation",
author = "Cree, {Michael J} and Olson, {John A} and McHardy, {Kenneth C} and Sharp, {Peter F} and Forrester, {John V}",
year = "1997",
month = "9",
doi = "10.1038/eye.1997.166",
language = "English",
volume = "11",
pages = "622--628",
journal = "Eye",
issn = "0950-222X",
publisher = "Nature Publishing Group",
number = "5",

}

TY - JOUR

T1 - A fully automated comparative microaneurysm digital detection system

AU - Cree, Michael J

AU - Olson, John A

AU - McHardy, Kenneth C

AU - Sharp, Peter F

AU - Forrester, John V

PY - 1997/9

Y1 - 1997/9

N2 - A fully automated digital image processing system, which provides an objective and repeatable way to quantify microaneurysms in digitised fluorescein angiograms, has been developed. The automated computer processing includes registration of same-eye retinal images for serial studies, cutting of regions-of-interest centred on the fovea, the detection of microaneurysms and the comparison of serial images for microaneurysm turnover. The microaneurysm detector was trained against a database of 68 images of patients with diabetes containing 394 true microaneurysms, as identified by an ophthalmologist. The microaneurysm detector achieved 82% sensitivity with 2.0 false-positives per image. An independent test set, comprising 20 images containing 297 true microaneurysms, was used to compare the microaneurysm detector with clinicians. The microaneurysm detector achieved a sensitivity of 82% for 5.7 false-positives per image, whereas the clinician receiver-operator-characteristic (ROC) curve gives 3.2 false-positives per image at a sensitivity of 82%. It is concluded that the computer system can reliably detect microaneurysms. The advantages of the computer system include objectivity, repeatability, speed and full automation.

AB - A fully automated digital image processing system, which provides an objective and repeatable way to quantify microaneurysms in digitised fluorescein angiograms, has been developed. The automated computer processing includes registration of same-eye retinal images for serial studies, cutting of regions-of-interest centred on the fovea, the detection of microaneurysms and the comparison of serial images for microaneurysm turnover. The microaneurysm detector was trained against a database of 68 images of patients with diabetes containing 394 true microaneurysms, as identified by an ophthalmologist. The microaneurysm detector achieved 82% sensitivity with 2.0 false-positives per image. An independent test set, comprising 20 images containing 297 true microaneurysms, was used to compare the microaneurysm detector with clinicians. The microaneurysm detector achieved a sensitivity of 82% for 5.7 false-positives per image, whereas the clinician receiver-operator-characteristic (ROC) curve gives 3.2 false-positives per image at a sensitivity of 82%. It is concluded that the computer system can reliably detect microaneurysms. The advantages of the computer system include objectivity, repeatability, speed and full automation.

KW - Aneurysm

KW - Diabetic Retinopathy

KW - Disease Progression

KW - False Positive Reactions

KW - Fluorescein Angiography

KW - Follow-Up Studies

KW - Humans

KW - Image Interpretation, Computer-Assisted

KW - ROC Curve

KW - Sensitivity and Specificity

KW - Microaneurysm

KW - Diabetes

KW - Retinopathy

KW - Computer

KW - Digitisation

U2 - 10.1038/eye.1997.166

DO - 10.1038/eye.1997.166

M3 - Article

VL - 11

SP - 622

EP - 628

JO - Eye

JF - Eye

SN - 0950-222X

IS - 5

ER -