Border Identification of Echocardiograms via multiscale edge detection and shape modeling

An algorithm for endocardial and epicardial border iden-ti cation of the left ventricle in 2-D short-axis echocar-diographic images is presented. Our approach relies on shape modeling of endocardial and epicardial boundaries and prominent border information extracted from image sequences. The algorithm consists of four steps; wavelet-based edge detection, border segment extraction, border reconstruction, and boundary smoothing. Wavelet maxi-mum representation of edges, dynamic shape modeling and matched ltering techniques are utilized to determine the center point of the left ventricle, and carry out feature ex-traction of border segments to better approximate endocar-dial and epicardial boundaries. The algorithm can reliably estimate the center point of the left ventricle, and also de-termine both endocardial and epicardial boundaries. My-ocardial boundary identication is autonomous requiring no human input for initial estimation of boundary locations. Sample experimental results are shown for endocardial and epicardial border identication in 2-D short-axis echocar-diograms. 1