Cataract and Glaucoma Disease Detection Using Deep Learning

Date
2022
Student ID
Editor
Journal Title
Volume
Issue
Journal Title
Journal ISSN
Volume Title
Abstract
Classification of cataract and Glaucoma is critical for evaluating eye disease and making treatment decisions based on their classes. Due to its remarkable performance, deep learning based image classification has been used to detect cataract and Glaucoma disease. Fundus images are commonly used for those classifications due to their superior image quality. The objective of this project is to classify binary class eye disease by using a deep, convoluted neural network custom model. The network is trained and tested on T1-weighted contrast-enhanced images. The performance of the network was evaluated using 10-fold cross-validation methods, and the improvement of the network was tested by using augmented images. The proposed network performs significantly better, with an overall average accuracy of 99%. Without cross validation, the accuracy of the model is 98%. Moreover, the record-wise cross-validation for the augmented data set obtained the best result for the 10-fold cross-validation method, with an average accuracy of 99%. The results show that the model may be used to classify cataract and glaucoma disease very precisely. Overall, this model can be used to classify the eye disease very efficiently and with less cost because of its outstanding performance.
Description
Keywords
Citation
Department Name
Electrical and Computer Engineering
Publisher
North South University
Printed Thesis
DOI
ISSN
ISBN