An Experimental Approach Pertaining To Transfer Learning Models For Classification Of Diabetic Retinopathy

One of the most prevalent eye conditions and a diabetes side effect that impacts vision is diabetic retinopathy. There may be no or very few signs of diabetic retinopathy. Blindness might eventually result. So early detection of symptoms may help prevent blindness. The features of exudates, blood vessels, and microaneurysms-three elements of diabetic retinopathy-are the subject of several studies that we discuss in this study. We may classify distinct stages into healthy, slightly non-proliferative, moderately non-proliferative, severely non-proliferative, and proliferative stages using these features. The stages of diabetic retinopathy were predicted using the following techniques: convolutional neural networks, inception v3, and efficient net. The best results came from CNN and Inception V3.