Retinal vessel segmentation in colour fundus images using Extreme Learning Machine

The retinal vessels are the only part of the blood circulation system of human that can be non-invasive observed directly (Resnikoff et al., 2004; Mehriban and Chen, 2012). Many systemic conditions such as diabetes, hypertension, arteriosclerosis, and cardiovascular disease can be found by the retinal vasculature examination. Therefore, the attributes of the retinal vessel such as width, tortuosity, branching pattern and angles play significant role in early disease diagnosis. Furthermore, various pathological changes of retinal vessels are direct reflections of fundus disease, such as diabetic retinopathy (DR), glaucoma, age-related macular degeneration (AMD) and so on.

Source: Computerized Medical Imaging and Graphics - May 30, 2016 Category: Radiology Authors: Chengzhang Zhu, Beiji Zou, Rongchang Zhao, Jinkai Cui, Xuanchu Duan, Zailiang Chen, Yixiong Liang Source Type: research