Automated system for the detection of hypertensive retinopathy

Retinal abnormality associated with high blood pressure is known as hypertensive retinopathy (HR). This disease leads to permanent blindness so timely diagnosis and treatment of the disease is very important. Fundus image analysis is used to diagnose HR. An automated system for the early detection of HR proves to be useful for ophthalmologists and patients. In this paper, we propose an automated system for the detection of HR using arteriovenous (AV) ratio. The proposed system consists of novel method for classification of vessels as arteries and veins using new feature vector and hybrid classifier. This paper also presents a new method to calculate vessel width which is useful to measure AV ratio. The system detects whether fundus image contains HR or not using already calculated AV ratio. Two publicly available digital fundus image databases i.e. VICAVR and DRIVE are used for the testing of our proposed algorithm. Experimental results show the validity of our proposed algorithm.