Velocity-based cardiac segmentation and motion-tracking.

摘要

A velocity-aided cardiac segmentation method based a modified active contour model, the orientation gradient force, and phase contrast MRI has been developed to improve the accuracy of segmentation of the myocardial boundaries, especially the endocardial boundary. Furthermore, the initial seed contour tracking algorithm was also developed to improve the accuracy of automatic sequential frame segmentation. Experimental results showed modest improvements in individual frame segmentation and noticeable in sequential frame segmentation.; A cardiac motion tracking method based on elastic deformation estimation of a deformable model and phase contrast MRI has been also developed to track the three-dimensional motion of the myocardium addressing the problem of relatively low out-of-plane resolution of MRI. To assess the proposed method, the motion of a normal human left ventricle was tracked throughout the entire cardiac cycle. Then, the tracking results were compared with a color-coded tissue Doppler echocardiography data set. A strain analysis was also carried out. The results showed that the out-of-plane velocity measurements from the model were well correlated with those of the echocardiography data set at the selected regions in the septum and the strain measurements were generally found to be consistent with previously published values.