A Novel Deep Ensemble Learning Approach to Predict Eye Diseases from OCT Images

As the world advances in technology, the classification of eye diseases is thought to have clinical applications since it evaluates and analyzes the results. Optometrists research and diagnose age-related eye problems such as Choroidal Neovascularization (CNV), Diabetic Macular Edema (DME), and Drusen using optical coherence tomography (OCT) of the human eye. Because their symptoms are similar, identifying these diseases is challenging and requires hours of expert analysis. The requirements for eye illness classification should be straightforward in nature, structure, and evaluation. Deep Learning Convolutional Neural Network Classification can create a hierarchical representation of images to differentiate between normal and diseased eye patterns for diagnostic purposes. Therefore, this paper aims to implement and define deep learning to classify and identify eye diseases. Using a four-class classification problem, this work investigates the automatic detection of Choroidal Neovascularization (CNV), Diabetic Macular Edema (DME), Drusen, and Normal in Optical Coherence Tomography (OCT) pictures. Multi-transfer learning architecture was used in the suggested deep learning Ensemble classification model. The most incredible accuracy of our proposed Ensemble model for multi- class classification was 83%.