Optimizing diabetic retinopathy detection with inception-V4 and dynamic version of snow leopard optimization algorithm

Diabetic retinopathy is a severe ocular condition that can result in vision loss due to damage to the retinal vessels. Early detection is of paramount importance in reducing the risk of further vision impairment and guiding appropriate treatment strategies. This study presents an innovative approach to enhance the accuracy and efficiency of diabetic retinopathy detection by integrating the Inception-V4 deep learning-based neural network with a modified dynamic Snow Leopard Optimization (DSLO) algorithm. The DSLO algorithm optimizes feature selection, thereby contributing to improved diagnostic performance. By analyzing digital images obtained during routine eye exams, automated image processing algorithms can identify early signs of diabetic retinopathy, such as leaking vessels or optic nerve edema. The proposed Inception-V4/DSLO model is evaluated using a practical dataset, Diabetic Retinopathy 2015, and compared to other state-of-the-art models, including mining local and long‐range dependence (MLLD), parallel convolutional neural network (PCNN) and ELM classifier (PCNN/ELM), diabetic retinopathy using convolutional neural networks for feature extraction and classification (DRFEC), Retrained AlexNet convolutional neural network (R-AlexNet), and Deep-DR demonstrating superior performance and improved detection of early-stage diabetic retinopathy cases.