Diabetic Retinopathy Detection using Advanced Deep Learning Algorithms

Khokale, Mrunali Narayan (2023) Diabetic Retinopathy Detection using Advanced Deep Learning Algorithms. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (3MB) | Preview
Preview	PDF (Configuration Manual) Download (1MB) | Preview

Abstract

Current methods of diabetic retinopathy involve manual inspection by trained professionals, which is time-consuming and requires expertise. This study aims to detect diabetic retinopathy, a major cause of blindness in people with diabetes, using advanced deep-learning algorithms. The focus of the study is on creating and testing a Transformer-CNN hybrid model against well-known architectures such as EfficientNet, ResNet50, InceptionNet, Efficient Net. Models were trained and tested using the Diabetic Retinopathy Comptetion Dataset, which included retinal images with different levels of retinopathy, to see how well these models compare based on accuracy. The results show that the ResNet50 model did better than the others with an accuracy of 80%, indicating its prowess in the classification of complex retinal images. Transformer-CNN, EfficientNet, and InceptionNet, on the other hand, had similar accuracy rates of about 73%. Selecting the right deep learning architectures for specific medical imaging tasks is very important, as shown by this variation in accuracies. The novelty of the study lies in the implementation of the Transformer-CNN hybrid model. The model did not perform significantly better than traditional architectures in this application but achieved comparable accuracy to other models, maybe owing to its smaller architecture failing to extract deep features. Research shows that advanced deep-learning techniques could be used to detect diabetic retinopathy early. This could lead to better diagnostic procedures and better patient care in the field of ophthalmology.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Qayum, Abdul UNSPECIFIED
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science R Medicine > RE Ophthalmology Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	14 May 2025 13:53
Last Modified:	14 May 2025 13:53
URI:	https://norma.ncirl.ie/id/eprint/7549

Actions (login required)

View Item