A Robust Approach based on CNN-LSTM Network for the identification of diabetic retinopathy from Fundus Images

Diabetic retinopathy (DR) is a condition that infects the eyes that involve people with diabetes losing their vision. It influences the blood vessels of the eye. Sometimes people complain about eyesight problems, such as difficulties reading or seeing too far away. The retina's blood vessels begin to bleed in the later disease later stages. Highly trained experts typically examine coloured fundus images to detect this fatal condition. This condition's manual diagnosis is time-consuming and error-prone. As a result, many computers vision-based algorithms for automatically detecting DR and its various stages from retina images have been presented. We used the Kaggle retina image dataset for this study, which is openly accessible. We introduced a convolutional neural network (CNN), and long short-term memory (LSTM) based deep learning technique for diagnosing diabetic retinopathy. The system attains better performance than the existing systems.