Automated identification of the heart wall in echocardiographic images throughout a cardiac cycle

In most of the methods for evaluation of cardiac function based on echocardiography, the heart wall, which is the object to be analyzed, is currently identified manually by an operator. However, this task is very time-consuming and suffers from inter- and intraobserver variability. In the present study, a method was proposed for automated identification of the heart wall region in ultrasonic data throughout an entire cardiac cycle. The heart wall region at the optimal frame of interest in the cardiac phase was identified by applying the expectation-maximization (EM) algorithm to extracted features, and heart wall regions in the following frames were identified by tracking each point, which was classified in the optimal (initial) frame as the heart wall region, using the phased tracking method. The results showed that the proposed method was feasible in automated identification of the heart wall in the longitudinal axis view.