Coronary movement analysis using X-ray cineangiographic images

Presents a methodology for finding point correspondences in non rigid motion of coronary arteries using cineangiographic images. It assumes that the coronary contour can not change its shape very rapidly. Points of correspondence on vessels are estimated using two successive frames based on minimization of deformation measurements between the curves. The cost function considers the displacement vector of corresponding matched points and curvature information. The search strategy for optimal matching is carried out using dynamic programming. Results of the method applied to real and synthetic images have shown that the proposed approach is accurate, with a root mean square error less than one pixel for simulated data. For the case of actual X-ray coronary angiographic images, visual inspection of the detected pairs of points shows that the results are very encouraging