In recent years there has been an increasing interest in quantitative analysis of coronary cineangiograms and already for a longer time of left ventricular cin- eangiograms. The needfor quantitationofcoronary arterialdimensions has been stimulated by the introduction ofnew therapeutic procedures in the catheteriza- tionlaboratory, suchas the balloon dilatationtechnique (PTCA) and thromboly- tic therapy, by the need to study the vasoactive responses of pharmaceutical agents, and also by the desire to study the progressive nature ofcoronary artery disease with the ultimate goal to find ways to bring a halt to the progression of coronary atherosclerosis or even achieve regression of the disease. Parallel with these clinical developments, rapid technical developments in computerarchitect- ures and semiconductor memories have made it possible to digitize and store cineframesor selected portions thereof in image processors and to analyze these pictorial data quantitatively at affordable prices. More than 15 years of research have been directed by various groups towards the semi- or fully-automated delineation of the left ventricular boundaries on a frame-to-frame basis.
Yet not a single system with fully-automated capability is commercially available. In the mean time many different left ventricular wall motion models have been developed, again with little consensuson which model is to be preferred as no golden standard exists.