NORMA eResearch @NCI Library

Diabetic Retinopathy stages Classifications using Lesion features and CNN Models

Anil Kumar, Sangeetha Pillay (2022) Diabetic Retinopathy stages Classifications using Lesion features and CNN Models. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
PDF (Master of Science)
Download (1MB) | Preview
[thumbnail of Configuration manual]
PDF (Configuration manual)
Download (1MB) | Preview


It is generally known that diabetic retinopathy (DR) is one of the most prevalent global causes of vision loss in people between the ages of 25 to 74. Because (DR) may initially cause no symptoms or only very minor vision problems, predicting DR in the early stages is crucial. The ultimate aim of this study is to classify the stage of DR such as mild, moderate, proliferate and severe. This study makes use of two CNN models; the first will determine whether there is DR present or not, and the second will classify the DR picture according to its stage. The model made use of the lesion feature to aid in this classification. This was compared with a model that does not classify the stages of DR using the lesion feature (DR Stage Classification). In 10 epochs, the VGG-16 model of the DR stage classification model provided training accuracy of 56.77% and test or validation accuracy of 54.73%, whereas the VGG-19 model produced training accuracy of 60.28% and test or validation accuracy of 54.04%. And the Lesion Feature DR Stage Classification model outperformed both the VGG-16 and VGG-19 models, with training accuracy of 90.88% and test or validation accuracy of 52.03 % and training accuracy of 95.29% and test or validation accuracy of 48.65% in 250 epochs, respectively.

Item Type: Thesis (Masters)
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
R Medicine > R Medicine (General)
R Medicine > RB Pathology
R Medicine > RE Ophthalmology
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Tamara Malone
Date Deposited: 17 Jan 2023 16:58
Last Modified: 17 Jan 2023 16:58

Actions (login required)

View Item View Item