Determining heart parameters through left ventricular automatic segmentation for heart disease diagnosis

Segmentation of the left ventricle in echocardiographic images is a task with important diagnostic power. This paper describes a method for the calculation of the ventricle’s contours in long- and short-axis views. The segmentation process starts with a complex border detection process, followed by the application of a generalized Hough transform to detect curves and ended up through an active-contour algorithm. Afterwards, parameters left ventricular area in a short-axis view, left ventricular length in a long-axis view, left ventricular volume, left ventricular mass, and heart-wall thickness are specifically calculated.