Polynomial Regression Based Edge Filtering for Left Ventricle Tracking in 3D Echocardiography
Automated detection of endocardial borders in 3D echocardiography is a challenging task. Part of the reason for this is the endocardial boundary leads to alternating edge characteristics that vary over a cardiac cycle. The maximum gradient (MG), step criterion (STEP) and max flow/min cut (MFMC) edge detectors have been previously applied for the endocardial edge detection problem. In this paper, a local polynomial regression based method (LPR) is introduced for filtering the STEP results. For each endocardial model point, (1) the surface is parametrized locally around the point, (2) a polynomial regression is applied on the STEP edges in the parametric domain, and (3) the fitted polynomial is evaluated at the origin of the parametric domain to determine the endocardial edge position. The effectiveness of the introduced method is validated via comparative analyses among the MFMC, STEP, and first & second degree LPR methods.
KeywordsPolynomial Regression Parametric Domain Surface Error Step Edge Endocardial Border
Unable to display preview. Download preview PDF.
- 1.Yang, L., Georgescu, B., Zheng, Y., Meer, P., Comaniciu, D.: 3d ultrasound tracking of the left ventricles using one-step forward prediction and data fusion of collaborative trackers. In: Proceedings of IEEE Conf. Computer Vision and Pattern Recognition (2008)Google Scholar
- 2.Orderud, F., Rabben, S.I.: Real-time 3d segmentation of the left ventricle using deformable subdivision surfaces. In: Proceedings of IEEE Conf. Computer Vision and Pattern Recognition (2008)Google Scholar
- 8.Dikici, E., Orderud, F.: Graph-cut based edge detection for kalman filter based left ventricle tracking in 3d+ t echocardiography. In: Proceedings of Computing in Cardiology (2010)Google Scholar