Classification of blood vessels as arteries and veins for diagnosis of hypertensive retinopathy

Vasculature abnormalities are the indicator of different diseases in the human body. Retinal blood vessels are very sensitive to blood pressure changes. Diameter abnormalities of retinal blood vessels are the first clinical finding in many retinal diseases such as glaucoma, Diabetic retinopathy, hypertensive retinopathy and macular degeneration. Automated and accurate classification of blood vessels into arteries and veins may help the ophthalmologist to find the retinal disorders. In this paper, we present a novel method for automated detection of hypertensive retinopathy. The proposed system classifies the vessel into arteries and veins using different machine learning techniques and then detects hypertensive retinopathy by computing arteriolar to Venular ratio. The proposed system is tested on one publicly available database and one locally gathered database. The quantitative results show the validity of proposed system.