Hybrid Fuzzy Assisted RNN Algorithm for Diabetic Retinopathy Identification

The most common cause of blindness in diabetics is diabetes retinopathy. The importance of early detection in preventing the DR cannot be overstated. However, because eye checkups are so expensive, many people let DR worsen and eventually lead to blindness. To identify persons with DR, the current study employs a color fundus image and a Fuzzy RNN algorithm. Although the purpose of this study is to save money, it could be a game changer for people with DR who cannot afford a medical diagnosis. Prior processing methods include green band extraction, histogram equalization, filtration, optic disc removal with organizational components on architecturally procedures, and illumination changes. The data acquired by GLCM contains contrast, correlation, energy, and homogeneity, and it is utilised to extract characteristics from the preprocessing results. This paper describes a threshold segmentation-based DR detection system. As a result of this study, a fuzzy-RNN methodology for identifying diabetic retinopathy has been developed. Threshold-based categorization is used to determine the diabetic zone. GLCM (Grey Level Co-occurrence Matrix) is the feature extraction technique applied in this case. These characteristics are utilised for categorization. For classification, an RNN classifier is utilised. To simulate the image processing using Phython in diabetic retinopathy is used to validate the final outcome.