Enhancing Diabetic Retinopathy Detection using Hybrid Deep Learning Approach

One of the main causes of impaired vision is diabetic retinopathy or DR, for which rapid and precise diagnosis is essential to successful medical care. This inquiry presents a hybridized deep learning AI model that combines a Support Vector Machine, that's the SVM classifier for improved outlined elegance with MobileNet for extracting the features. With low computational cost, MobileNet, a light weighted convolutional neural network, accurately extracts significant spatial information from retinal fundus imagery. SVM is also used to classify these detached characteristics, which improves the model's capacity to distinguish across different DR phases. In comparison to existing deep learning approaches, the suggested method delivers increased precision, excellence, and learned by fusing SVM's powerful framed features with Mobile Net’s crucial feature description.