Detecting diabetic retinopathy exudates in fundus images using fuzzy c-means (FCM)

Diabetic Retinopathy (DR) is the main cause of blindness for diabetic patients. As the exudates are the primary sign of DR, therefore early detection and timely treatment can prevent and delay the risk of vision loss. Automatic screening could facilitate the screening process, reduce inspection time...

Full description

Saved in:
Bibliographic Details
Main Authors: Tahreer Dwickat, Hadi Hamad
Format: Article
Language:English
Published: An-Najah National University 2020-10-01
Series:مجلة جامعة النجاح للأبحاث العلوم الطبيعية
Subjects:
Online Access:https://journals.najah.edu/media/journals/full_texts/3_4mBZdHr.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850039003897659392
author Tahreer Dwickat
Hadi Hamad
author_facet Tahreer Dwickat
Hadi Hamad
author_sort Tahreer Dwickat
collection DOAJ
description Diabetic Retinopathy (DR) is the main cause of blindness for diabetic patients. As the exudates are the primary sign of DR, therefore early detection and timely treatment can prevent and delay the risk of vision loss. Automatic screening could facilitate the screening process, reduce inspection time, and increase accuracy, which is vital in ophthalmic treatment, this development of exudates detection will help doctors in detecting symptoms faster. In this research, we use an automatic method to detect exudates from retinal digital images with non-dilated pupils of retinopathy patients; starting by detecting both the optic disc (OD) and retinal vessels, then probable exudates are defined through morphological techniques, in the last main phase, four features are implemented as input data for the fuzzy C-means (FCM) clustering to define the existing exudates in the fundus images. The overall detection performance is evaluated through measuring sensitivity, specificity, positive predictive value (PPV), positive likelihood ratio (PLR) and accuracy. These measures are done by comparing results to hand-drawn ground truth (GT) done by an expert; which are comparatively analyzed. It is found that the proposed method detects exudates successfully with average values of sensitivity, specificity, PPV, PLR and accuracy of 86.3%, 98.4%, 20.8%, 86.2 and 98.4% respectively on the testing studied database.
format Article
id doaj-art-dbb83d55c2474ed2a8f60226b260341a
institution DOAJ
issn 1727-2114
2311-8865
language English
publishDate 2020-10-01
publisher An-Najah National University
record_format Article
series مجلة جامعة النجاح للأبحاث العلوم الطبيعية
spelling doaj-art-dbb83d55c2474ed2a8f60226b260341a2025-08-20T02:56:27ZengAn-Najah National Universityمجلة جامعة النجاح للأبحاث العلوم الطبيعية1727-21142311-88652020-10-01351376410.35552/anujr.a.35.1.1861Detecting diabetic retinopathy exudates in fundus images using fuzzy c-means (FCM)Tahreer Dwickat0Hadi Hamad1MA student: Department of Mathematics, Faculty of Science, An-Najah National University, Nablus, PalestineDepartment of Mathematics, Faculty of Science, An-Najah National University, Nablus, Palestine.Diabetic Retinopathy (DR) is the main cause of blindness for diabetic patients. As the exudates are the primary sign of DR, therefore early detection and timely treatment can prevent and delay the risk of vision loss. Automatic screening could facilitate the screening process, reduce inspection time, and increase accuracy, which is vital in ophthalmic treatment, this development of exudates detection will help doctors in detecting symptoms faster. In this research, we use an automatic method to detect exudates from retinal digital images with non-dilated pupils of retinopathy patients; starting by detecting both the optic disc (OD) and retinal vessels, then probable exudates are defined through morphological techniques, in the last main phase, four features are implemented as input data for the fuzzy C-means (FCM) clustering to define the existing exudates in the fundus images. The overall detection performance is evaluated through measuring sensitivity, specificity, positive predictive value (PPV), positive likelihood ratio (PLR) and accuracy. These measures are done by comparing results to hand-drawn ground truth (GT) done by an expert; which are comparatively analyzed. It is found that the proposed method detects exudates successfully with average values of sensitivity, specificity, PPV, PLR and accuracy of 86.3%, 98.4%, 20.8%, 86.2 and 98.4% respectively on the testing studied database.https://journals.najah.edu/media/journals/full_texts/3_4mBZdHr.pdfdiabetic retinopathyfcm.retinaexudates
spellingShingle Tahreer Dwickat
Hadi Hamad
Detecting diabetic retinopathy exudates in fundus images using fuzzy c-means (FCM)
مجلة جامعة النجاح للأبحاث العلوم الطبيعية
diabetic retinopathy
fcm.
retina
exudates
title Detecting diabetic retinopathy exudates in fundus images using fuzzy c-means (FCM)
title_full Detecting diabetic retinopathy exudates in fundus images using fuzzy c-means (FCM)
title_fullStr Detecting diabetic retinopathy exudates in fundus images using fuzzy c-means (FCM)
title_full_unstemmed Detecting diabetic retinopathy exudates in fundus images using fuzzy c-means (FCM)
title_short Detecting diabetic retinopathy exudates in fundus images using fuzzy c-means (FCM)
title_sort detecting diabetic retinopathy exudates in fundus images using fuzzy c means fcm
topic diabetic retinopathy
fcm.
retina
exudates
url https://journals.najah.edu/media/journals/full_texts/3_4mBZdHr.pdf
work_keys_str_mv AT tahreerdwickat detectingdiabeticretinopathyexudatesinfundusimagesusingfuzzycmeansfcm
AT hadihamad detectingdiabeticretinopathyexudatesinfundusimagesusingfuzzycmeansfcm