Major temporal arcade separation in angiography images of retina using the Hough transform and connected components

Abstract

Structural analysis of the vascular architecture of the retina is useful in facilitate the diagnosis of retinopathy disease. Retinopathy can affects the blood flow velocity. Measuring blood flow velocity in major temporal arcade (MTA) vessels in retina is one of the approaches to identify blood flow velocity changes. Hence separation of MTA vessel is a necessary step in blood flow velocity computation. We propose methods for separation of the MTA. We use Gabor filters to detect retinal vessels and Hough transform to model the MTA. Afterwards we use the MTA model to find initial point of MTA. Next we take advantage of connected components to track temporal arcade vessel, and separate the MTA. Results obtained with applying our algorithm to 24 clinical angiography images, in which traces of MTA was drawn with an ophthalmologist expert, indicating 92% correct separation of MTA.