Multiple Eye Disease Detection using Hybrid Adaptive Mutation Swarm Optimization and RNN

The major cause of visual impairment in aged people is due to age related eye diseases such as cataract, diabetic retinopathy, and glaucoma. Early detection necessary for better diagnosis. This paper concentrates on the early identification various disorders glaucoma from retinal fundus images. proposed method focuses automated multiple using hybrid adaptive mutation swarm optimization regression neural networks (AED-HSR). In work, input images are preprocessed then features entropy, mean, color, intensity, standard deviation, statistics extracted collected data. segmented by an (AMSO) algorithm segment disease sector image. Finally, fed a network (RNN) classifier classify each image normal or abnormal. If output abnormal, it classified corresponding terms glaucoma, which improves accuracy classification. Ultimately, results classifiers evaluated several performance analyses viability structural functional considered. system predicts type with 0.9808, specificity 0.9934, sensitivity 0.9803 F1 score 0.9861 respectively.