Graph Tracer ? Simultaneous Vessel Identification to Detect the Retinal Crossover Points.

Retinal vessel morphology is highly related with hypertension, coronary heart disease, and stroke and the changes in retinal vessel identification are considered as earlier diagnosis of the cardiovascular disease. However, there is a difficulty in quantifying changes in retinal vessels. Accurate measurement of vessel diameters on retinal images plays an important part in diagnosing cardiovascular diseases. In order to diagnose the diseases correctly, a method of vessel diameter measurement has been developed with a tracking technique for simultaneous identification of true vessels. Hence post processing step vascular structure segmentation is introduced for identifying the true vessels. So the vessel segment graph is constructed and formulated as the problem of finding optimal forest in the graph. Graph Tracer Algorithm computes the problem as constraint optimization problem (COP) to identify all the crossover points and segments. It represents the vessels in a binary tree and searches for an optimal forest. Thus simultaneous identification of vessels gives good diagnosis. Proposed work includes cup and OD segmentation for glaucoma assessment which prevents from visual loss with earlier treatment.