Classification of Retinal Diseases Using Transfer Learning Approach

Retinal disease classification of Color Fundus Images using Computational Imaging has been widely accepted. This paper proposes a transfer learning approach for Retinal Disease Classification using Deep Learning. Deep learning methods are widely used due to higher performance and accuracy but at the cost of time. A large amount of time and processing power is required to train a dataset and achieve considerable accuracy. Transfer learning reduces a load of using huge data-sets, thereby reducing the time complexity of the training data. We have selected VGG19 model to fine-tune the network for the extraction of knowledge. The VGG19 model is fine-tuned with a retinal database. The results contain accuracy the network on changing various hyper-parameters like learning rate, number of epochs, optimizer algorithm to find a suitable set of parameters for higher accuracy. The motivation for using Deep Neural network is the fact that it has multi-fold benefits like better performance compared to traditional machine learning techniques, higher accuracy, automated feature extraction and easier to design and process.