Eye Diseases Prediction and Classification using Deep Learning Techniques

In human anatomy, the eye is a major sense organ and loss of vision would have an enormous impact on quality of life. In addition, it may also be possible that the eye is showing signs of severe health problems. Eye diseases detection is mainly the problem here. Detecting eye diseases using different advanced AI- based techniques to analyze and predict the type of eye disease from retinal eye images and improve the performance and the accuracy of the proposed system. This proposed system can help ophthalmologists to save their time in manual examination. This paper aims to develop Deep Learning (DL) for prediction and classification of eye diseases by applying Gabor Filter as feature extractor as it can extract the important features such as textual and imaginary patterns of the structure and shape of the images and then use its output as an input to the Convolutional Neural Network (CNN) training models. Then classifying retinal eye images to predicate the type of the disease. After training and evaluating these trained models, the model with the best accuracy was VGG-16 with a model accuracy of 85.34%.