Minimally invasive epicardial activation mapping from multielectrode catheters

Catheter based electrophysiologic studies of the epicardium are limited to regions near the coronary arteries and veins or require transthoracic access. We have developed a statistical approach by which to estimate high resolution maps of epicardial activation from multielectrode venous catheter measurements. An essential component of this approach is the computation of an estimation matrix based on a training set of previously measured maps of epicardial activation. We report here on a strategy by which to improve estimation accuracy by matching the composition of the training set to features of the estimated activation map. Results suggest that matching based on the site of ectopic activation achieves an improvement in estimation accuracy. A further finding is that estimation accuracy was preserved when based on as few as 14 measurement electrodes, thus supporting the clinical feasibility of activation mapping from multielectrode catheters.