Abstract
Optical Coherence Tomography (OCT) is a vital imaging technique that provides detailed images of the retina, playing a crucial role in diagnosing and monitoring various retinal conditions like diabetic macular edema (DME), choroidal neovascularization (CNV), and DRUSEN. However, there is a need to improve early detection and treatment of these common eye diseases. While deep learning methods have demonstrated superior accuracy in analyzing OCT images especially concerning data volume and computational efficiency requires further exploration. This paper presents a comprehensive approach for classifying Optical Coherence Tomography (OCT) images using model ensemble. An ensemble model refers to the merging or blending of separate deep learning models, aiming to utilize their unique strengths and abilities to construct a more resilient and effective solution. The methodology involves the use of CNN architecture along with DenseNet121 and InceptionV3 models to enhance the accuracy of classifying retinal images into four categories: CNV, DME, DRUSEN, and NORMAL. By leveraging the strengths of these models, the proposed ensemble method achieves superior performance. The results demonstrate the effectiveness of the ensemble approach, with an improvement in classification accuracy compared to individual models. The proposed architecture achieved the accuracy of 97.5%. The performance comparison with existing state-of-the-art techniques demonstrates that the proposed algorithm requires significantly less time with limited dataset. Our proposed method shows the performance of OCT classification in the case of a limited dataset.