Diabetic Macular Edema (DME) is the most common cause of permanent vision loss among people with diabetic retinopathy. However, early detection and treatment can reduce the risk of blindness. This paper presents an automatic method to detect DME and DR and oversteps the subjective manual evaluation of opthalmologists. Based on Convolutional Neural Network, a proposed end-to-end CNN model is presented and fully trained for the automatic classification of Optical Coherence Tomography (OCT) retinal imaging. The experiments over two datasets, provided by different institutions, have been evaluated by randomly shuffling and separating the training data along with test data. Using the proposed model, the experiment results showed a classification accuracy, sensitivity and specificity of 99.02. © 2018 IEEE.