Detection of Hypertension Retinopathy and Diabetes Using Machine Learning

Abstract

Medical conditions can be identified with only photographs of the eye which is called fundus images using computer automated processes. The task of distinguishing blood vessels by simply observing retinal images has proven to be challenging without the aid of technology. Various morphological features such as retinal veins, arteries, diameter, branching angle and length have diagnostic features. The tiny blood vessels can deteriorate and leak fluid into and under the retina which causes swelling of the retina. Hypertension Retinopathy has vascular damage caused by hypertension. Funduscopic examination shows cotton-wool spots, arteriolar construction, vascular wall changes, arteriovenous nicking, flame shaped hemorrhages, optic disc edema and yellow hard exudates.

Keywords: Fundus Photography, Retina, Eye, Hypertension Retinopathy, Diabetic Retinopathy, Hard Exudates, Cotton Wool