Retina Capturing Using Portable Device and Classification of Images for Prevalence of Eye Disease

In the healthcare field, the treatment of diseases is more effective when detected at an early stage, a lot of attention from researchers due to its far-reaching benefits and also decreases the load on doctors. There is a need for a dataset that is collected from portable devices which enables researchers to develop the research for screening of eye diseases. There are many ways available to capture fundus images and store them. Among all these devices few cannot store images of fundus, few can take the only video of fundus and few need an IOS device to capture images of the fundus. So, this proposal aims at a solution by designing a portable device that acts as a camera to capture fundus Images. The output video of the portable device is given as the input for the model that is to be developed for the creation of an image of fundus. This makes it easy for the creation of collection of a dataset of fundus which can further be used for training different models related to eye diseases. This project mainly focused on developing the device by considering limitations of existing devices and working on making a device to capture the retinal image. And the images generated by the device on applying data augmentation technique, which is GAN, helps to generate new samples from the existing image dataset collected from the device and classification of images for Diabetic Retinopathy eye disease.