Advanced Image Analysis for the Assessment of Retinal Vascular Changes

In the last decade, one of the major advances in retinal vascular imaging research has been the clear demonstration that physiological and pathological alterations in the retinal vascular network are associated with a variety of worldwide major diseases such as diabetes, hypertension and atherosclerosis. However, the clinical assessment of the retinal vascular condition is most of the times tiresome, and prone to errors, particularly if occurs in a screening environment. Recent advances in image analysis can avoid this workload and provide the ophthalmologist with objective and reproducible results useful in daily clinical practice. The retinal image analysis systems have therefore become a prominent and powerful diagnostic tools in the field of ophthalmology by detecting the changes in retinal images. It has been widely demonstrated that in diabetic retinopathy, the blood vessels often show abnormalities at early stages, as well as vessel diameter alterations. Changes in retinal blood vessels, such as significant dilatation and elongation of main arteries, veins, and their branches, are also frequently associated with hypertension and other cardiovascular pathologies. Among several characteristic signs associated with vascular changes, the Central Retinal Arteriolar Equivalent (CRAE), the Central Retinal Venular Equivalent (CRVE) and the Arteriolar-to-Venular Ratio (AVR) have been frequently used as indicators for the early detection, diagnosis, staging and follow-up of diabetes and hypertension, since they can reflect the narrowing or dilation of the retinal blood vessels. The main goal of this work is the development of an automatic system for the measurement of CRAE, CRVE, AVR and several bifurcation geometrical features. Among other image processing operations, the estimation of these features requires vessel segmentation, vessel caliber measurement, artery/vein (A/V) classification and optic disc (OD) segmentation.