Computer-aided diagnosis system for the assessment of retinal vascular changes

This paper presents an automatic application that provides several retinal image analysis functionalities, namely vessel segmentation, vessel width estimation, artery/vein classification and optic disc segmentation. A pipeline of these methods allows the computation of important vessel related indexes, namely the Central Retinal Arteriolar Equivalent (CRAE), Central Retinal Venular Equivalent (CRVE) and Arteriolar-to-Venular Ratio (AVR), as well as various geometrical features associated with vessel bifurcations. The results for AVR estimation were assessed using the images of INSPIRE-AVR dataset; for this dataset, the mean error of the measured AVR values with respect to the reference was identical to the one achieved by a medical expert. The estimation of the CRAE, CRVE and AVR values on 480 images from 120 subjects have shown a significant correlation between right and left eyes and also between images of same eye acquired with different camera fields of view.