Binary Classification of Fundus Images Using G-EYE for Disease Detection

Cataract, age-related macular degeneration, glaucoma, diabetic retinopathy, corneal opacity, and trachoma are the most common conditions that impair vision. Regarding the accessibility, cost, and level of population eye care literacy, there are significant differences in the reasons between and within nations. An effective classification model is required to raise the bar for eye care services and to enable hassle-free early detection of the most prevalent causes of vision impairment. A suitable and sophisticated dataset is necessary to do this. The major challenge that is encountered while working in this field is the scarcity of data (It is not easy to collect labelled data from ophthalmologist). In order to solve this issue, the G-EYE portable smartphone-based retinal imaging system was created. For clinical assessment, G-EYE can capture and transmit high-definition photographs and videos of the fundus. An efficient dataset of fundus images may be produced using the G-EYE device, and the images are then subjected to various image pre-processing techniques, including image sharpening and masking techniques, which can be used to classify the eye disease.