Original Research Article Fast, accurate and robust retinal vessel segmentation system Zhexin Jiang Q1 , Juan Yepez, Sen An, Seokbum Ko * University of Saskatchewan, Department of Electrical and Computer Engineering, 57 Campus Drive, Saskatoon, Canada S7N 5A9 1. Introduction Q2 The retinal vasculature has been acknowledged as an indispensable element in both ophthalmological and cardio- vascular disease diagnosis such as glaucoma and diabetic retinopathy. The attributes of retinal blood vessels including length, width, tortuosity, branching pattern and angles will contribute to the diagnostic result. However, manual segmen- tation of retinal blood vessels, although possible, is a time consuming and repetitive work, and it also requires profes- sional skills. To assist ophthalmologists with this complex and tedious work, the demand for the fast automated analysis of the retinal vessel images arises. However, to fully automate the analysis and make it work for real-life diagnosis is a harsh task. First, because even the thinnest vessel may contribute to the differential diagnosis list, in order to avoid medical accidents, the extraction of blood vessels is required to be extremely accurate so as to help the diagnosis. Second, for many diseases such as diabetes and hypertension, patients are required to take regular ocular screening in order to detect retinopathy in early stages. However, patients who are inconvenient to move or live distantly from the city will be less approachable for the location-specific treatment. In this case, the automated solution is expected to be handy and portable in the future. Besides, almost every retinal vasculature extraction system is facing the trade-off between accuracy and computing b i o c y b e r n e t i c s a n d b i o m e d i c a l e n g i n e e r i n g x x x ( 2 0 1 7 ) x x x – x x x 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 * Corresponding author at: University of Saskatchewan, Department of Electrical and Computer Engineering, 57 Campus Drive, Saskatoon, Canada S7N 5A9. a r t i c l e i n f o Article history: Received 13 December 2016 Accepted 9 April 2017 Available online xxx Keywords: Retinal images Vessel segmentation Morphological processing DoOG filter Automated analysis Thresholding a b s t r a c t The accurate segmentation of the retinal vessel tree has become the prerequisite step for automatic ophthalmological and cardiovascular diagnosis systems. Aside from accuracy, robustness and processing speed are also considered crucial for medical purposes. In order to meet those requirements, this work presents a novel approach to extract blood vessels from the retinal fundus, by using morphology-based global thresholding to draw the retinal venule structure and centerline detection method for capillaries. The proposed system is tested on DRIVE and STARE databases and has an average accuracy of 95.88% for single- database test and 95.27% for the cross-database test. Meanwhile, the system is designed to minimize the computing complexity and processes multiple independent procedures in parallel, thus having an execution time of 1.677 s per image on CPU platform. © 2017 Published by Elsevier B.V. on behalf of Nalecz Institute of Biocybernetics and Biomedical Engineering of the Polish Academy of Sciences. BBE 195 1–10 Please cite this article in press as: Jiang Z, et al. Fast, accurate and robust retinal vessel segmentation system. Biocybern Biomed Eng (2017), http://dx.doi.org/10.1016/j.bbe.2017.04.001 Available online at www.sciencedirect.com ScienceDirect journal homepage: www.elsevier.com/locate/bbe http://dx.doi.org/10.1016/j.bbe.2017.04.001 0208-5216/© 2017 Published by Elsevier B.V. on behalf of Nalecz Institute of Biocybernetics and Biomedical Engineering of the Polish Academy of Sciences.