Chronic eye disease diagnosis using ensemble-based classifier

In this paper, we address the problem of diagnosis of a chronic eye disease which is the Primary Open Angle Glaucoma (POAG) disease. It is called the “sneak thief” of sight since the slowly and undetected damage of the optic nerve in the early stages of the disease. The research tends to achieve an early and accurate diagnosis of glaucoma based on an ensemble classifier by integrating the principal component analysis (PCA) with rotation forest tree (ROT). The evaluation performance has been accomplished with three well known classifiers namely Neural Network (NN), Decision Tree (DT) and Fuzzy Logic classifiers. The data is collected using an empirical data collected of Egypt Air hospital. It was observed that ROT achieved the highest classification accuracy in most tested cases. The experiments show that the overall accuracy offered by the employed technique is high compared to other machine learning techniques used in the research. It is believed that the ensemble classifier ROT could perform high accuracy in classification process.