Glaucoma stages detection using feature selection techniques

Glaucoma, also known as the "silent thief of sight", is a chronic eye disease caused by increased intraocular pressure, leading to irreversible vision loss without the affected individual ever being aware of it. Over time, it results in damage to the optic nerve, which carries visual images from the eye to the brain. It slowly destroys the eyes of the patients unsuspectingly. There are no early symptoms of glaucoma. Therefore, the challenge lies in detecting the early stages of the disease. This paper puts forth a methodology of feature extraction and relevant feature selection. Feature selection is a much important Stage because it cuts the time required for training time and the features extracted by selecting only worthy and important features while neglecting all unnecessary or non-important features that helps in classifier boosting accuracy. ANN were the methods used in classifying. The result obtained from the proposed method was evaluated with accuracy, specificity and sensitivity. The experimental results of the proposed system achieved an accuracy of 96.12 % with the use of the Harvard dataset and 91.67 % using the Rim-One1 dataset. The experimental proved that the proposed model outperform advanced methods in diagnosing early stages of glaucoma.