Fusion of Vision Transformer, Inception-V3 and ResNet50 for Efficient Eye Disease Detection

Abstract

The rising prevalence of retinal disorders globally is a major challenge for the identification and classification of retinal diseases in the medical system. Timely diagnosis and treatment of disorders like glaucoma, DR, and macular degeneration are essential for preventing irreversible vision loss. Better patient outcomes and less strain on healthcare systems can result from the significant improvement in the precision and effectiveness of identifying a variety of ocular diseases by the utilization of Artificial Intelligence (AI), Machine Learning (ML) algorithms, and modern imaging technologies. The Vision Transfer (ViT) Algorithm, a novel method for the accurate identification and categorization of ocular disorders, is presented in this work. With standard datasets, our algorithm performs exceptionally well in classifying various eye diseases. Through the integration of sophisticated image processing methods with ML, the ViT Algorithm demonstrates strong performance in differentiating between various eye conditions. The outcomes demonstrate how well it works to increase diagnostic precision and make quick interventions possible. By providing an efficient method for reliable and rapid disease identification, this research significantly advances the field of ocular healthcare and improves patient outcomes.