Acquisition, segmentation and tracking of the cerebral vascular tree on 3Dmagnetic resonance angiography images

This paper presents a method for the detection, representation and visualis ation of the cerebral vascular tree and its application to magnetic resonan ce angiography (MRA) images. The detection method is an iterative tracking of the vessel centreline with subvoxel accuracy and precise orientation est imation. This tracking algorithm deals with forks. Centrelines of the vesse ls are modelled by second-order B-spline. This method is used to obtain a h igh-level description of the whole vascular network. Applications to real a ngiographic data are presented. An MRA sequence has been designed, and a gl obal segmentation of the whole vascular tree is realised in three steps. Ap plications of this work are accurate 3D representation of the vessel centre line and of the vascular tree, and visualisation. The tracking process is a lso successfully applied to 3D contrast enhanced MR digital subtracted angi ography (3D-CE-MRA) of the inferior member vessels. In addition, detection of artery stenosis for routine clinical use is possible due to the high pre cision of the tracking algorithm. (C) 2001 Elsevier Science B.V. All rights reserved.