Intensity Ridge and Widths for Tubular Object Segmentation and Description
MMBIA '96 Proceedings of the 1996 Workshop on Mathematical Methods in Biomedical Image Analysis (MMBIA '96)
Registration and Analysis of Vascular Images
International Journal of Computer Vision - Special Issue on Research at the University of North Carolina Medical Image Display Analysis Group (MIDAG)
A review of vessel extraction techniques and algorithms
ACM Computing Surveys (CSUR)
Machine Graphics & Vision International Journal
A Statistical Approach for Detecting Tubular Structures in Myocardial Infarct Scars
FIMH '09 Proceedings of the 5th International Conference on Functional Imaging and Modeling of the Heart
| Hi-index | 0.00 |
We present a new metric for registering 3D images of vasculature, and we analyze the rigid-body transformation parameter space of that metric and its derivatives. To quantify and direct a source image's alignment with a target image, this new vascular-image registration system models the vessels in the source image and makes measurements in the target image at a sparse set of transformed points from the centerlines of those models. The system is fast and effective because the measures made at the transformed centerline points incorporate the general geometric properties of tubes and specific model-quality information calculated during the vessel model generation process. Additionally, by adjusting the sample density or scaling the centerline point measures, coarse-to-fine registration strategies are directly enabled. We present visualizations of the metrica nd its derivatives over a range of mis-registrations given different sample densities and different measure scalings using magnetic resonance angiograms, x-ray computed tomography images, and 3D ultrasound images.