A Review on Swarm intelligence & Evolutionary Algorithms based Approaches for Diabetic Retinopathy Detection

Diabetic retinopathy has overtaken cataracts as the primary cause of new blindness globally. Diabetics are more likely to develop cataracts, visual loss, glaucoma, excessive intraocular pressure, and, most importantly, diabetic retinopathy (DR). If blood vessels in the retina are compromised, vision loss is irreversible. The patient may not exhibit any symptoms early on, and by the time they do, the damage has already been done. Early diabetes treatment helps to retain vision and permits a patient to see. Diabetic retinopathy is a worldwide health problem. To address the medical community’s requests for early identification of diabetes and other illnesses, several professionals have advocated a computer assisted diagnosis technique. In this work, image processing techniques and image classifiers that sort images based on the status of the disease will be used to describe automated ways to look at retinal images for important signs of diabetic retinopathy. There are compelling motivations to create retinopathy risk reduction models and strategies that can be used widely. The difficulty of acquiring accurate diabetic retinopathy at a reasonable cost needs a major investment in creating and testing computer-assisted diagnosis (CAD). This study looks at the different stages, traits, and types of models that may be used to reduce the risk of diabetic retinopathy and detect it early using Evolutionary computing and Swarm optimization.