Echocardiographic definition of the left ventricular centroid. I. Analysis of methods for centroid calculation from a single tomogram.

Quantitation of myocardial contraction requires a frame of reference. Most investigators have sought a single reference frame per image, centered in some manner with respect to the mass of myocardium. Because there is no anatomic marker for the center of the heart, many different approaches have been pursued to identify a centroid of the left ventricle. The issue of whether the reference should be fixed throughout the cardiac cycle or float from image to image has been addressed in previous studies, but the more fundamental question of how a centroid can best be defined has not been answered. This study examines this basic issue by analysis of variance from observer to observer, cycle to cycle, animal to animal and method to method. Both endocardial and epicardial borders were digitized twice by each of two observers at 1/30 s intervals spanning the cardiac cycle for each of three cardiac cycles in six normal dogs. The left ventricular centroid was calculated by six methods: center of endocardial coordinates, center of epicardial coordinates, center of mid-myocardial (average) coordinates, center of endocardial area, center of epicardial area and center of mid-myocardial (average) area. The path of each centroid was correlated between observers and correlation coefficients were transformed for analysis of variance. This analysis indicates a best approach to centroid definition through distinct minimization of the variance: the best of the six methods proved to be center of endocardial area.