Automatic extraction of cerebral arteries from magnetic resonance angiography data: Algorithm and validation Suhuai Luo T , Sungyiau Lee, Xin Ma, Aamer Aziz, Wieslaw L. Nowinski Biomedical Imaging Lab, Agency for Science Technology and Research, Singapore Abstract. We present a cerebral vasculature extraction method from magnetic resonance angiography (MRA) and provide validation for arteries. After reviewing the state-of-the-art in vasculature segmentation techniques, we introduce an automatic algorithm with robust maximal intensity searching and region growing. We present the details of the design and application of extraction validation interface. We demonstrate the artery extraction fidelity of the method with tests on both 3D phantom and real MRA images. We conclude by summarising the proposed algorithm and pointing out possible future pursuits in vessel segmentation. D 2005 Published by Elsevier B.V. Keywords: Extraction; Segmentation; MRA; Vasculature; Artery 1. Introduction Automatic cerebral vasculature extraction is needed to reveal 3D vessel structure from MRA data. This will make possible to automatically detect an early stage of vascular diseases, such as carotid stenosis, aneurysm, and vascular malformation. Other useful applications of automatic vasculature segmentation include: vessel tree registration with an atlas, virtual endoscopy, and vasculature visualization. In segmentation of cerebral vessels from MRA data, special considerations should be given to the complexity of the data, including (a) complicated vascular tree, consisting of hundreds of branches; (b) curvy and tortuous vessels; (c) the topology and morphology of the vasculature varying according to different subjects and sampling times; (d) low intensity contrast between a vessel and its 0531-5131/ D 2005 Published by Elsevier B.V. doi:10.1016/j.ics.2005.03.276 T Corresponding author. E-mail addresses: suhuai@bii.a-star.edu.sg, suhuailuo@optusnet.com.au (S. Luo). International Congress Series 1281 (2005) 375 – 380 www.ics-elsevier.com