New techniques for the processing and analysis of retinal images in diagnostic ophtalmology

This thesis deals with the automatic analysis of color fundus images and with its application to diagnostic ophthalmology. Diabetes is a growing epidemia in the world, due to population growth, aging, urbanization and increasing prevalence of obesity and physical inactivity, so diabetic retinopathy has an ever increasing importance as a cause of blindness. Also hypertension affects the microcirculation and hypertensive retinopathy is one of the consequences of such damage. In this thesis new algorithms to help ophthalmologist's diagnosis and to be used in automated systems for retinopathy screening will be presented .