A Boolean characterization of three-dimensional simple points
Pattern Recognition Letters
Liver Blood Vessels Extraction by a 3-D Topological Approach
MICCAI '99 Proceedings of the Second International Conference on Medical Image Computing and Computer-Assisted Intervention
IPMI '93 Proceedings of the 13th International Conference on Information Processing in Medical Imaging
A review of vessel extraction techniques and algorithms
ACM Computing Surveys (CSUR)
ISVC'05 Proceedings of the First international conference on Advances in Visual Computing
A review on MR vascular image processing: skeleton versus nonskeleton approaches: part II
IEEE Transactions on Information Technology in Biomedicine
Segmentation of 4D cardiac MRI: Automated method based on spatio-temporal watershed cuts
Image and Vision Computing
Hi-index | 0.00 |
Magnetic resonance angiography (MRA) provides three-dimensional data of vascular structures by visualising the flowing blood signal. Algorithms dedicated to vessel segmentation generally detect the cerebral vascular tree by only seeking this high intensity blood signal in MRA data. The method presented in this paper proposes a different strategy which consists in using both MRA and classical MRI in order to integrate a priori anatomical knowledge for guidance of the vessel segmentation process. It then uses mathematical morphology tools to carry out a simultaneous segmentation of both blood signal in MRA and blood and wall signal in MRI, enabling to take advantage of a larger amount of information than previously proposed methods. This method is dedicated to the superior sagittal sinus segmentation; however, similar strategies could be considered for segmentation of other vascular structures. It has been performed on a database composed of 9 couples of MRA and MRI, providing results which have been validated and compared to other ones obtained with a region-growing algorithm. Their validation tends to prove that the proposed method is reliable even when the vascular signal is inhomogeneous or contains artifacts.