Automatic vessel extraction by patient motion correction and bone removal in brain CT angiography

Abstract

We propose a robust 3D rigid registration and masking techniques for getting a clear arterial anatomy and easily detecting cerebral aneurysms, arterial stenosis and other vascular anomalies in brain CT angiography. Our method is composed of the following steps. First, a set of feature points are automatically selected using a 2D edge detection technique within the ring model. Second, a Gaussian-weighted distance map is constructed for leading our similarity measure to robust convergence on the optimal value. Third, the similarity measure between feature points is evaluated repeatedly by weighted cross-correlation. Fourth, bone-vessel masking and subtraction is performed for completely removing bones. Our method has been successfully applied to 10 patients of pre- and post-contrast images of brain CT angiography. Experimental results show that the performance of our method is very promising compared with conventional methods in the aspects of its visual inspection, accuracy and robustness. Our method can be useful for the early diagnosis and treatment planning of cerebrovascular anomalies in CT angiography.