NORMA eResearch @NCI Library

Detection of Diabetic Retinopathy using Deep Learning

Ghosh, Suvrojitmoy (2020) Detection of Diabetic Retinopathy using Deep Learning. Masters thesis, Dublin, National College of Ireland.

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

Abstract

Diabetic retinopathy is a disease caused due to untreated diabetes for a long period of time. This disease is increasing world-wide and is a major reason for blindness. The effective way of cure for this disease is to detect it at an early state. Even though there are different types of screening techniques but there is still scope for improvements. One of the way to improve the accuracy is by concentrating more on the factors present in the eye which helps in determining the stages of diabetic retinopathy, and secondly by using pre-trained deep convolution deep neural network which can reduce the cost of training. In this paper we have experimented with different types of pre-processing techniques and their effects on enhancing the model performance. Finally the enhanced images were classified using pre-trained models. Total six pre-trained models were used out of which three were fine-tuned namely ResNet-50, Inception-v3 and InceptionResNet-v2 and the rest were used as conventional transfer learning models namely Inception-v4, DenseNet-169 and DenseNet-201. A brief comparison was drawn between the fine-tuned models and the conventional models, using the evaluation metrics like accuracy, validation accuracy and model loss. From the above mentioned experimentation we can conclude that preprocessed images along with fine-tuned pre-trained models are the best combination for diabetic retinopathy classification.

Item Type: Thesis (Masters)
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
Q Science > QA Mathematics > Computer software
T Technology > T Technology (General) > Information Technology > Computer software
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Dan English
Date Deposited: 22 Jan 2021 12:35
Last Modified: 22 Jan 2021 12:35
URI: https://norma.ncirl.ie/id/eprint/4442

Actions (login required)

View Item View Item