Detection of lesions and classification of diabetic retinopathy using fundus images

Simple item page
dc.contributor.authorPaing, May Phu
dc.contributor.authorChoomchuay, Somsak
dc.contributor.authorRapeeporn Yodprom, M. D.
dc.date.accessioned2026-08-06T10:16:14Z
dc.date.available2026-08-06T10:16:14Z
dc.date.issued2017-02-21
dc.description.abstractDiabetes retinopathy is a retinal disease that is affected by diabetes on the eyes. The main risk of the disease can lead to blindness. Detection the disease at early stage can rescue the patients from loss of vision. The major purpose of this paper is to automatically detect as well as to classify the severity of diabetic retinopathy. At first, the lesions on the retina especially blood vessels, exudates and microaneurysms are extracted. Features such as area, perimeter and count from these lesions are used to classify the stages of the disease by applying artificial neural network (ANN). We used 214 fundus images from DIARECTDB1 and local databases. We found that the system can give the classification accuracy of 96% and it supports a great help to ophthalmologists.
dc.identifier.citationBmeicon 2016 9th Biomedical Engineering International Conference, 2017
dc.identifier.doi10.1109/BMEiCON.2016.7859642
dc.identifier.other2-s2.0-85015876209
dc.identifier.urihttps://dspace.kmitl.ac.th/handle/123456789/7528
dc.sourceBmeicon 2016 9th Biomedical Engineering International Conference
dc.subjectblood vessels
dc.subjectdiabetic retinopathy
dc.subjectexudates
dc.subjectmicroaneurysms
dc.subjectneural network
dc.titleDetection of lesions and classification of diabetic retinopathy using fundus images
dc.typeConference Paper

Files

Collections

Research