ABSTRACT
An active microvascular condition of diabetes mellitus called diabetic retinopathy (DR) causes blurred vision, floaters, and difficulties seeing colors and sometimes results in permanent vision loss. Clement cases may be treated with cautious management of diabetes. Advanced cases may need surgery or laser treatment. So, early detection of DR is very important. The retinal fundus images are mainly used for the analysis and detection of DR, and this analysis is done by the Asia Pacific Tele-Ophthalmology Society 2019 Blindness Detection (APTOS 2019 BD) data set. There are a lot of DL algorithms available for utilization by anyone. The top ten most popular of them are CNN, recurrent neural networks, long short-term memory network, generative adversarial networks, radial basis function networks, self-organizing maps, multilayer perceptrons, deep belief networks, restricted Boltzmann machines, and Autoencoders. Future research will concentrate on hybrid-based strategies that will remove the drawbacks of the current retinal vascular extraction method.
