NORMA eResearch @NCI Library

Retinal Fundus Image Classification using LSTM - Convolution Neural Network

Vaydande, Rahul Suryakant (2022) Retinal Fundus Image Classification using LSTM - Convolution Neural Network. 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


In recent years, the prevalence of diabetes and its associated complication, diabetic retinopathy (DR), has increased alarmingly. Since DR might eventually result in permanent blindness, early detection is essential. As a result, countries undertake a variety of screening programs to prevent DR. Due to the scarcity of skilled professionals who can execute the diagnostic, DR identification remains a challenge. There is hence a clear need to create an automated DR detection system. Convolutional Neural Network (CNN) algorithms are the most often used DNN techniques in the field of medical image classification. The Convolution Neural Network in combination with a Recurrent Neural Networks (CNN-RNN) framework has demonstrated success in an Image classification task across different domains. The suggested hybrid CNN-RNN models combine the benefits of standard CNN with Long Short-Term Memory (LSTM). CNN models collect several distinct characteristics from fundus pictures, and indeed the retrieved features are categorised using LSTM. For this research, I have utilized the Kaggle Open-source dataset. The Kaggle dataset for DR suffers from a class imbalance problem. As a result, augmentation methods are utilized to normalize the dataset. The CNN-LSTM model which was developed produced a descent multi-class classification accuracy of 76% when compared to the pre-trained transfer learning models i.e., VGG19 and ResNet50 which produced accuracies of 75% and 47% 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 > RE Ophthalmology
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Tamara Malone
Date Deposited: 14 Mar 2023 12:38
Last Modified: 14 Mar 2023 12:38

Actions (login required)

View Item View Item