학술논문

Exploiting texture information in diagnosing Glaucoma
Document Type
Conference
Source
2017 25th Signal Processing and Communications Applications Conference (SIU) Signal Processing and Communications Applications Conference (SIU), 2017 25th. :1-4 May, 2017
Subject
Communication, Networking and Broadcast Technologies
Signal Processing and Analysis
Support vector machines
Retina
Histograms
Optical coherence tomography
Feature extraction
Three-dimensional displays
Entropy
Optical Coherence Tomography (OCT)
Gloucoma
Texture Analysis
Support Vector Machines (SVM)
Feature Selection
Classification
Language
Abstract
The most common cause of blindness in the world by far is known to be the Glaucoma condition. The increase in the ratio of cup to the disc area and the thinning of retinal layers are the most common symptoms of Glaucoma. Functional and structural features of the eye should be examined in order to distinguish an eye with Glaucoma from a healthy eye. In this study, the texture information in Optical Coherence Tomography (OCT) images, which is one of the functional features is investigated to get the most distinguishing characteristic texture patterns in retina layers. First, sample areas are extracted from 4 main region around the fovea using a systematic approach and calculated the texture features for each area one by one. After this step SVM classifier is exploited to find the features which can impact the diagnosis of Glaucoma condition. This analysis is useful to guide the scientists to diagnose glaucoma where the thickness information is not available or at the beginning of the Glaucoma when thinning is not started yet.