Automatic detection and tracking of left ventricular landmarks in echocardiography

We introduce a method for automatic detection and tracking of the apex and two landmarks defining the atrioventricular plane in apical views of the left ventricle. The method is based on using tissue Doppler to track a set of candidate points in a cardiac cycle. Each candidate point is given a score based on the gray-scale, the velocity, and the depth profiles. The landmarks with the lowest costs are selected, and measurements are calculated automatically. The method is evaluated by comparing with manual selection by four cardiologists, and the presented results show that the method is robust and extracted measurements are comparable with manual selection.