Human Eye Diagnosis for Cataract Detection from Funduscopic Images using Convolutional Neural Network

Date
2022
Research Supervisor
Editor
Journal Title
Volume
Issue
Journal Title
Journal ISSN
Volume Title
Abstract
Cataracts are one of the most common visual problems and it affects people as they grow older and lose their vision. A cataract is a cloud that forms on the lens of our eyes and it is caused by a buildup of debris. Among the most noticeable signs of this illness are blurred vision, faded colors and difficulties seeing in bright light. It is common for these symptoms to result in trouble doing a number of duties. Therefore, early cataract detection and prevention may aid in reducing the rate of blindness in the population. On the basis of a publicly available image dataset, the authors of this article hope to recognize cataract eye disease using convolutional neural networks. As part of this experiment, four alternative Convolutional Neural Network (CNN) meta-architectures, including NetInceptionV3, InceptionResnetV2, XceptionNet, MobileNet, EfficientNetB1, EfficientNetB3, EfficientNetV2B0 and EfficientNetV2B1 were applied to the TensorFlow object detection framework, with each architecture being represented by a different color.
Description
Keywords
TECHNOLOGY::Electrical engineering, electronics and photonics::Electrical engineering
Citation
Department Name
Electrical and Computer Engineering
Publisher
North South University
Printed Thesis
DOI
ISSN
ISBN