An Enhanced Technique To Classify OCT Images Using Deep Learning

In medical imaging field, computer-aided detection (CADe) or computer-aided diagnosis (CADx) is the computer-based system that helps doctors to take decisions swiftly. So, majority of doctors use this kind of CADs for the faster detection and diagnosis of ocular diseases. There are some major retinal diseases, they are Choroidal Neo-Vascularization (CNV), Drusen, and Diabetic Macular Edema (DME). These ocular diseases can result in partial or complete loss of vision. Optical Coherence Tomography (OCT) is widely used in diagnosis of these ocular diseases. In this paper, it have shown the comparison results of several network model and transfer learning on networks with pre-trained models was achieved. The dataset for the training and validation was taken from the Kaggle website which contains an approximate of 84.5k images. The custom-built sequential model has achieved more validation accuracy than other network models.