Automatic Location of Blood Vessel Bifurcations in Digital Eye Fundus Images
Retinal blood vessels are linked with hypertension and cardiovascular disease. It is generally known that vascular bifurcation is mainly involved in varying blood flow velocity as well as its pressure. This paper presents an efficient method for automatic
- PDF / 6,422,507 Bytes
- 11 Pages / 439.37 x 666.14 pts Page_size
- 89 Downloads / 215 Views
)
(
)
Biomechanical Electronics Research Group, Department of Mathematics and Computer Science, Liverpool Hope University, Hope Park, Liverpool, UK {chaicht,barretm,nagara}@hope.ac.uk 2 Department of Medical Radiation Sciences, School of Science, Curtin University, Perth, WA, Australia [email protected]
Abstract. Retinal blood vessels are linked with hypertension and cardiovascular disease. It is generally known that vascular bifurcation is mainly involved in varying blood flow velocity as well as its pressure. This paper presents an efficient method for automatic location of blood vessel bifurcations in digital eye fundus images. The proposed algorithm comprised of three main steps: image enhance‐ ment, fuzzy clustering, and searching vascular bifurcation. The purposed algo‐ rithm revealed successful detection of bifurcations upon test images. Results showed improved diagnostic accuracy in identifying bifurcations with use of the proposed algorithm and encourage its use for further applications such as image registration, personal identification and pre-clinical scanning of retina diagnosis. Keywords: Bifurcation · Retinal image · Fuzzy clustering · Imaging algorithm · Automatic location · Fundus image
1
Introduction
Bifurcation is a common connection in vascular network, basically a connected form of vascular tree structure in retinal image. Blood vessels within the eye supply the blood to retina. Fundus image is a photograph that captures the base of eyeball and principally provides basic information to characterise human eye condition. The main features in retinal image include blood vessels, optic nerve, macula and bifurcation [1–3]. In addi‐ tion, many eye diseases do not show any signs and symptoms, and they may be painless with no any change in the vision and it can be noticed until the condition is detected at an advanced stage. Retinal image analysis is an important research context to the devel‐ opment of computer-assisted diagnosis and biomedical imaging analysis due to a large amount of future data storage technology and clinical informatics [4]. It requires preclinical scanning software prototype to review fundus images for retina diagnosis that can provide quick information and feedback on retinal main features. Those details can be used to assist ophthalmologist in term of diagnosis and treatments for reducing time, © Springer Nature Singapore Pte Ltd. 2017 K. Deep et al. (eds.), Proceedings of Sixth International Conference on Soft Computing for Problem Solving, Advances in Intelligent Systems and Computing 547, DOI 10.1007/978-981-10-3325-4_33
Automatic Location of Blood Vessel Bifurcations
333
prevention, and making decision for further planning and or surgery. Retinal images representing healthy and eye disease are shown in Fig. 1. Previous works on identifying vascular bifurcations in fundus images have focused on characterising the vascular junctions as well as vascular crossover-sections. Fatepuria et al. [7] proposed the windows matching techniques to search locations of vascular cro
Data Loading...
