Accurate Classification of Tessellated Retinal Disease with Deep Learning Model

We provide a novel Tessellated Eye Classifier (TEC) model designed for accurate identification of retinal diseases, with a focus on Tessellation Disease (TSL). Our trials were carried out on the AWS cloud platform, utilizing cutting-edge technology such as optimized RAM and high-performance GPU computers. The TEC model is based on the VGG-16 architecture and has extra Convolutional layers, global average pooling, and a customized output layer designed for the categorization of normal circumstances and retinal disorders, such as TSL. To improve the performance of the model, training and validation were carried out using a variety of fundus images with consistent color channels and resolutions. After undergoing repeated training, the customized TEC model successfully classified retinal disorders with a notable test set accuracy of roughly 79%. This study provides insights into the complexity of the neural network by analyzing parameter distribution in both our proprietary TEC model and the VGG-16 base model in depth. These results highlight the importance of the TEC model for patient care and early diagnosis; more research is being done to optimize the model further to increase its accuracy and therapeutic usefulness.