Understanding and Visualization of Different Feature Extraction Processes in Glaucoma Detection

Abstract

In the recent years the usage of mobile phone is increased and it is the major reason for cause of vision loss in several people. The continuous usage increases pressure inside optic nerve head and it leads to glaucoma disease. Also, there are lot of other reasons which leads to the cause of glaucoma. The purpose of this paper is to determine the importance of feature extraction process in glaucoma detection and implementation of different techniques for extracting convenient features for training machine learning model using pre-processed OCT (Optical Coherence Tomography) images. The two major feature extraction techniques narrated in this paper are convolutional neural network (CNN) model-based feature extraction and image processing model-based feature extraction. A performance analysis was conducted to find best feature extraction technique and both techniques performed well.