Impact of image resolution on retinal vessel tortuosity assessment

Abstract

Several ocular and cardiovascular disorders are characterized by an elevated tortuosity of retinal blood vessels. Objective tortuosity measurements can be obtained through the computation of a set of tortuosity indices derived from digital image analysis of an eye fundus image. These calculations require the parametrization of the vessel path traced in the image. As a vessel descriptor, tortuosity should be independent of scale, but this property demands thorough examination and assessment for each index. We investigate the impact of altering image resolution on the local tortuosity indices of retinal vessels.