Glaucoma Detection with Fully Convolutional Neural Network using Optic Disc and Segmentation Methods

The disease which affects the optic nerve of an eye and also it leads to the cause of irreversible blindness globally named glaucoma. A vertical CDR of the fundus eye image is an important clinical indicator for glaucoma diagnosis. Optic Nerve Head (ONH) assessment method plays the major role for de...

Full description

Saved in:
Bibliographic Details
Main Authors: Shankar, B Maruthi, Nagaraj, V, Sivakumar, S A, Vidhya, B, Ganesh, R Senthil, Premalatha, P
Format: Conference Proceeding
Language:English
Subjects:
Computer architecture
Convolutional neural network
Cup-to-disc ratio
Deep learning
Glaucoma
Head
Image segmentation
Integrated optics
Optic disc &cup segmentation
Optic Nerve Head
Optical computing
Optical fiber networks
Training data
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The disease which affects the optic nerve of an eye and also it leads to the cause of irreversible blindness globally named glaucoma. A vertical CDR of the fundus eye image is an important clinical indicator for glaucoma diagnosis. Optic Nerve Head (ONH) assessment method plays the major role for detecting this kind of CDR used for ONH evaluation. It mainly focuses on noise removal process. Preprocessing of images is done for enhancing the contrast between cup and disk. For accurate measurement, the segmentation has been done automatically for separating the cup and disk. The Fully Convolutional Network (FCN) is employed as a main core in this deep neural network architecture. The transformation of fully connected layer into Convolutional layer is obtained by FCN and also the classification map of the image which has been sent of same size is done using upsampling.
ISSN:2575-7288
DOI:10.1109/ICACCS51430.2021.9441545