<p>Early detection and diagnosis of some diseases in the retina of the eye can improve the chances of cure and also prevent blindness. Retinal abnormalities have emerged as a serious public health concern in recent years and can manifest gradually and without warning. These diseases can affect any part of the retina, causing vision impairment and indeed blindness in extreme cases. This necessitates the development of automated approaches to detect retinal diseases more precisely and, preferably, earlier. In this paper a novel deep learning based opto- structural graphical memory network was presented for retinal disease prediction. Here initially the Optical Coherence Tomography (OCT) images are retrieved from Kaggle and the images were processed using the inherited speckle filter and data augmentation. Then the feature extraction step was carried out using the canny gauss Shi-Tomasi algorithm(CGSTA). Followed by feature extraction feature selection can be done using binary transport BAT optimization(BTBO) algorithm. Finally the selected features are given as a input to the opto- structural graphical memory network(OSGMN). The outcomes are then evaluated, and the efficiency of the novel strategy is contrasted with that of conventional approaches. Evaluations of the quantitative findings for precision, accuracy, recall and F score showed that they outperformed the state-of-the-art approaches.</p>