Automated microaneurysm detection using local contrast normalization and local vessel detection

Alan D Fleming, Keith A Goatman, John A Olson, Peter F Sharp, Sam Philip

Research output: Contribution to journalArticle

183 Citations (Scopus)

Abstract

Screening programs using retinal photography for the detection of diabetic eye disease are being introduced in the U.K. and elsewhere. Automatic grading of the images is being considered by health boards so that the human grading task is reduced. Microaneurysms (MAs) are the earliest sign of this disease and so are very important for classifying whether images show signs of retinopathy. This paper describes automatic methods for MA detection and shows how image contrast normalization can improve the ability to distinguish between MAs and other dots that occur on the retina. Various methods for contrast normalization are compared. Best results were obtained with a method that uses the watershed transform to derive a region that contains no vessels or other lesions. Dots within vessels are handled successfully using a local vessel detection technique. Results are presented for detection of individual MAs and for detection of images containing MAs. Images containing MAs are detected with sensitivity 85.4% and specificity 83.1%.

Original languageEnglish
Pages (from-to)1223-1232
Number of pages10
JournalIEEE Transactions on Medical Imaging
Volume25
Issue number9
DOIs
Publication statusPublished - Mar 2006

Keywords

  • Algorithms
  • Aneurysm
  • Artificial Intelligence
  • Diabetic Retinopathy
  • Humans
  • Image Enhancement
  • Image Interpretation, Computer-Assisted
  • Information Storage and Retrieval
  • Pattern Recognition, Automated
  • Reproducibility of Results
  • Retinal Vessels
  • Retinoscopy
  • Sensitivity and Specificity

Cite this

Automated microaneurysm detection using local contrast normalization and local vessel detection. / Fleming, Alan D; Goatman, Keith A; Olson, John A; Sharp, Peter F; Philip, Sam.

In: IEEE Transactions on Medical Imaging, Vol. 25, No. 9, 03.2006, p. 1223-1232.

Research output: Contribution to journalArticle

@article{793b571963e34189a1033f53599e1850,
title = "Automated microaneurysm detection using local contrast normalization and local vessel detection",
abstract = "Screening programs using retinal photography for the detection of diabetic eye disease are being introduced in the U.K. and elsewhere. Automatic grading of the images is being considered by health boards so that the human grading task is reduced. Microaneurysms (MAs) are the earliest sign of this disease and so are very important for classifying whether images show signs of retinopathy. This paper describes automatic methods for MA detection and shows how image contrast normalization can improve the ability to distinguish between MAs and other dots that occur on the retina. Various methods for contrast normalization are compared. Best results were obtained with a method that uses the watershed transform to derive a region that contains no vessels or other lesions. Dots within vessels are handled successfully using a local vessel detection technique. Results are presented for detection of individual MAs and for detection of images containing MAs. Images containing MAs are detected with sensitivity 85.4{\%} and specificity 83.1{\%}.",
keywords = "Algorithms, Aneurysm, Artificial Intelligence, Diabetic Retinopathy, Humans, Image Enhancement, Image Interpretation, Computer-Assisted, Information Storage and Retrieval, Pattern Recognition, Automated, Reproducibility of Results, Retinal Vessels, Retinoscopy, Sensitivity and Specificity",
author = "Fleming, {Alan D} and Goatman, {Keith A} and Olson, {John A} and Sharp, {Peter F} and Sam Philip",
year = "2006",
month = "3",
doi = "10.1109/TMI.2006.879953",
language = "English",
volume = "25",
pages = "1223--1232",
journal = "IEEE Transactions on Medical Imaging",
issn = "0278-0062",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "9",

}

TY - JOUR

T1 - Automated microaneurysm detection using local contrast normalization and local vessel detection

AU - Fleming, Alan D

AU - Goatman, Keith A

AU - Olson, John A

AU - Sharp, Peter F

AU - Philip, Sam

PY - 2006/3

Y1 - 2006/3

N2 - Screening programs using retinal photography for the detection of diabetic eye disease are being introduced in the U.K. and elsewhere. Automatic grading of the images is being considered by health boards so that the human grading task is reduced. Microaneurysms (MAs) are the earliest sign of this disease and so are very important for classifying whether images show signs of retinopathy. This paper describes automatic methods for MA detection and shows how image contrast normalization can improve the ability to distinguish between MAs and other dots that occur on the retina. Various methods for contrast normalization are compared. Best results were obtained with a method that uses the watershed transform to derive a region that contains no vessels or other lesions. Dots within vessels are handled successfully using a local vessel detection technique. Results are presented for detection of individual MAs and for detection of images containing MAs. Images containing MAs are detected with sensitivity 85.4% and specificity 83.1%.

AB - Screening programs using retinal photography for the detection of diabetic eye disease are being introduced in the U.K. and elsewhere. Automatic grading of the images is being considered by health boards so that the human grading task is reduced. Microaneurysms (MAs) are the earliest sign of this disease and so are very important for classifying whether images show signs of retinopathy. This paper describes automatic methods for MA detection and shows how image contrast normalization can improve the ability to distinguish between MAs and other dots that occur on the retina. Various methods for contrast normalization are compared. Best results were obtained with a method that uses the watershed transform to derive a region that contains no vessels or other lesions. Dots within vessels are handled successfully using a local vessel detection technique. Results are presented for detection of individual MAs and for detection of images containing MAs. Images containing MAs are detected with sensitivity 85.4% and specificity 83.1%.

KW - Algorithms

KW - Aneurysm

KW - Artificial Intelligence

KW - Diabetic Retinopathy

KW - Humans

KW - Image Enhancement

KW - Image Interpretation, Computer-Assisted

KW - Information Storage and Retrieval

KW - Pattern Recognition, Automated

KW - Reproducibility of Results

KW - Retinal Vessels

KW - Retinoscopy

KW - Sensitivity and Specificity

U2 - 10.1109/TMI.2006.879953

DO - 10.1109/TMI.2006.879953

M3 - Article

VL - 25

SP - 1223

EP - 1232

JO - IEEE Transactions on Medical Imaging

JF - IEEE Transactions on Medical Imaging

SN - 0278-0062

IS - 9

ER -