Performance Analysis of Diabetic Retinopathy Using Deep Learning Algorithm

Blindness can result from the disease known as Diabetic Retinopathy (DR), which damages retinal blood vessels. Typically, fundus photos in various colors are utilized to identify this incurable illness. Therefore, a variety of computer vision practical engineering techniques are used to automatically forecast the DR's occurrences and stages. However, due to their high processing costs and inability to extract highly nonlinear characteristics, these techniques are unable to accurately categorize the various stages of DR. The proposed Visual Geometry Group (VGG) model can handle a DR image at any size and tends to better classification. In this paper, we have used VGG architecture for DR image classification. In this research work, two retinal image open access databases have been involved with the annotation of retinal images by the expert eye specialist. The databases used in this research work are DIARETDB1 and SYSU. Based on the unique features extracted, the suggested classifier had accurately detected the type of lesion that was available in the fundus image with an accuracy rate of 98%. The experimental findings demonstrate that, when compared to cutting-edge techniques, the suggested model outperforms them in terms of accuracy and computational resource usage.