OcuVision:CNN powered Analysis of Retinal images for Disease Diagnosis

In numerous areas of the world, retinal disorders are a major factor in vision loss and blindness. Early recognition and detection of various conditions, including diabetic retinopathy,macular degeneration, and glaucoma, are essential for avoiding irreversible vision loss.We intend to develop a multi-class retinal illness diagnosis method utilizing deep CNNs that is effective. Convolutional neural networks (CNNs) are utilized by the suggested EyeDeep-Net multi-layer neural network to extract pertinent information from fundus images and make diagnostic assessments. The technology holds promise for utilization in telemedicine scenarios,healthcare settings and point-of-care gadgets with limited resources, enabling the prompt identification of diseases and rapid intervention. The accuracy of the model is evaluated using multiple statistical criteria, showing better performance in both disease diagnosis and classification when compared to many modern baseline models. In order to train and assess the suggested deep learning model, this study makes use of the RFMiD dataset, a useful resource. What sets this work apart is the systematic analysis of different optimizers applied to the EyeDeep-Net architecture, which aims to enhance the model's performance and accuracy. The optimization process is vital in tailoring the model to achieve superior results in retinal disease diagnosis.