Point-tracked quantitative analysis of left ventricular surface motion from 3-D image sequences

We propose and validate the hypothesis that we can use differential shape p roperties of the myocardial surfaces to recover dense field motion from sta ndard three-dimensional (3-D) image sequences (MRI and CT). Quantitative me asures of left ventricular regional function can be Further inferred from t he point correspondence maps. The noninvasive, algorithm-derived results ar e validated on two levels. First, the motion trajectories are compared to t hose of implanted imaging-opaque markers of a canine model in two imaging m odalities, where subpixel accuracy is achieved. Second, the validity of usi ng motion parameters (path length and thickness changes) for detecting myoc ardial injury area is tested by comparing algorithms derived results to pos tmortem analysis TTC staining of myocardial tissue, where the achieved Pear son product-moment correlation value is 0.968.