Retinal diseases become more complex for people if they are not identified in the early stages. Diagnosing retinal diseases is time-consuming by using traditional approaches. Early detection of retinal diseases can prevent permanent loss. Optical coherence tomography (OCT) images are grey-scale images that show the back of the human eye which is called the retina. Deep Learning (DL) is most widely used to solve various issues that are more complex. In this paper, anEnsemble classification (EC) is introduced to detect and diagnose retinal diseases by using OCT images. EC is the combination of noise filters that removes the noise from given OCT input images and Convolutional Neural Network (CNN). Preprocessing of OCT images can be done by noise filters. The CNN is the DL model that can classify the OCT images according to the diseases. To improve the performance of EC the edge detection approach is adopted to find the accurate edges of the OCT image. Thus this can helps the experts to find the abnormalities present in the OCT input image. This paper focused on classifying the 4 types of diseases as Diabetic Macular Edema (DME), choroidal neovascularization (CNV), Drusen, Age-Related Macular Degeneration (ARMD), and normal case. The pre-trained model VGG-19 is used to train the model with the selected OCT images dataset. The performance of EC is analyzed by using a confusion matrix.

Ensemble Classification (EC), Deep Learning (DL), Diabetic Macular Edema (DME), choroidal neovascularization (CNV), Drusen, Diabetic Retinopathy (DR).