Identification of Human Eye Diseases Using GLCM and DNN

Abstract

Diseases such as cardiovascular diseases and tumors play a severe role in human health, and hence early and proper diagnosis is indispensable. Traditional diagnostic methods primarily consist of manual interpretation, which is time-consuming and prone to error. For overcoming these limitations, this research proposes an automatic disease detection system based on fundus images. The aim is to promote diagnostic efficiency and accuracy via advanced image processing and deep learning algorithms. The proposed system begins with preprocessing of the fundus images to remove noise and unwanted background characteristics. The color and texture characteristics are next extracted using the Grey-Level Co-Occurrence Matrix (GLCM) method. The characteristics are subsequently fed into a Deep Neural Network (DNN) model for classification and identification of the diseases. This procedure eliminates excessive dependence on human examination and enhances consistency in diagnosis. The system is a high-quality, scalable, and affordable way of assisting healthcare professionals and improving patient outcomes with early detection.