A Comparative Study of Retinal Vasculature Extraction in Digital Fundus Images

Abstract

Some of the most common blinding conditions are caused by choroidal neovascularization (CNV). The relevant conditions include diabetic retinopathy and age-related macular degeneration. At present, the only proven modality of effective treatment is the application of laser energy to the CNV to cauterize the vessels. The key to effective and lasting treatment is the identification of the full extent of the CNV, complete cauterization of the CNV by accurately aiming an appropriate amount of optical energy while ensuring that healthy tissue is not cauterized. Extraction techniques must be developed to discern the retinal blood vessels tree and determine the positions of laser shots in a reference frame. This paper presents an efficient comparison of different methods to segment blood vessels, which is a prominent anatomical structure in retina, in both gray-scale and color retinal images. The blood vessel extraction is composed of six algorithms according to two criteria, i.e., Extraction of the blood vessel boundaries (using Difference operators, Decision based-directional edge detection, Morphological gradient and Deformable model algorithm) & Extraction of the core area of the blood vessel tree by tracing vessels centers (using 2-dimensional matching filters and Morphological reconstruction algorithm). Results on various retinal images verify the effectiveness of the proposed methods.