J Anal Res Clin Med. 2016;4(2): 104-109.
doi: 10.15171/jarcm.2016.017
  Abstract View: 1802
  PDF Download: 2449

Original Article

Automatic detection of retinal exudates in fundus images of diabetic retinopathy patients

Mahsa Partovi 1, Seyed Hossein Rasta 2*, Alireza Javadzadeh 3

1 Department of Medical Physics, School of Medicine, Tabriz University of Medical Sciences, ‎Tabriz, Iran
2 Assistant Professor, Stem Cell Research Centre AND Department of Medical ‎Bioengineering, Tabriz University of Medical Sciences, Tabriz, Iran AND Honorary ‎Researcher, School of Medical Sciences, University of Aberdeen, Aberdeen, UK‎
3 Professor, Department of Ophthalmology, School of Medicine, Tabriz University of Medical ‎Sciences, Tabriz, Iran


Introduction: Diabetic retinopathy (DR) is the most frequent microvascular complication of diabetes and can lead to several retinal abnormalities including microaneurysms, exudates, dot and blot hemorrhages, and cotton wool spots. Automated early detection of these abnormalities could limit the severity of the disease and assist ophthalmologists in investigating and treating the disease more efficiently. Segmentation of retinal image features provides the basis for automated assessment. In this study, exudates lesion on retinopathy retinal images was segmented by different image processing techniques. The objective of this study is detection of the exudates regions on retinal images of retinopathy patients by different image processing techniques. Methods: A total of 30 color images from retinopathy patients were selected for this study. The images were taken by Topcon TRC-50 IX mydriatic camera and saves with TIFF format with a resolution of 500 × 752 pixels. The morphological function was applied on intensity components of hue saturation intensity (HSI) space. To detect the exudates regions, thresholding was performed on all images and the exudates region was segmented. To optimize the detection efficiency, the binary morphological functions were applied. Finally, the exudates regions were quantified and evaluated for further statistical purposes. Results: The average of sensitivity of 76%, specificity of 98%, and accuracy of 97% was obtained. Conclusion: The results showed that our approach can identify the exudate regions in retinopathy images.
Keywords: Automatic Detection, Hard Exudates, Diabetic Retinopathy, Image Processing, Color Fundus Imaging, Diagnostic Imaging
First Name
Last Name
Email Address
Security code

Abstract View: 1803

Your browser does not support the canvas element.

PDF Download: 2449

Your browser does not support the canvas element.

Submitted: 19 Jan 2016
Accepted: 14 Feb 2016
ePublished: 09 May 2016
EndNote EndNote

(Enw Format - Win & Mac)

BibTeX BibTeX

(Bib Format - Win & Mac)

Bookends Bookends

(Ris Format - Mac only)

EasyBib EasyBib

(Ris Format - Win & Mac)

Medlars Medlars

(Txt Format - Win & Mac)

Mendeley Web Mendeley Web
Mendeley Mendeley

(Ris Format - Win & Mac)

Papers Papers

(Ris Format - Win & Mac)

ProCite ProCite

(Ris Format - Win & Mac)

Reference Manager Reference Manager

(Ris Format - Win only)

Refworks Refworks

(Refworks Format - Win & Mac)

Zotero Zotero

(Ris Format - Firefox Plugin)