Detection of Retinal Features for Diagnosis of Diabetic Retinopathy

Human Retina has potential to reveal important information about retinal and ophthalmic diseases. Retinal imaging is being analyzed for the study of different diseases over the years. It has a stream of featured blood vessels. Change or modification in these blood vessels can trigger the abnormalities that could lead to a certain disease while the disease is still in its earliest stage. A large scale screening implementation is generally required for identifying any victim of these diseases that involves efforts in acquisition of the retinal images and examining these images under different benchmarks. There was a need to research and made the system for better detecting the symptoms for different ophthalmic disease victim. The declared system for detection of diabetic retinopathy and macular degeneration had been intake the insight from all of those feature detection approaches and delivered full screening system for these diseases. The screening system involved image acquisition, noise removal, image centering, region selection, vessel segmentation, feature extraction, and classification. The system has been first implemented on some publically available datasets and then looked for the operational capabilities of the system by running some clinical diagnosis over it.