The inverse problem of electrocardiography in terms of epicardial potentials and their gradients

We have explored an approach to the inverse problem of electrocardiography that yields, in addition to the electric potentials on the epicardial surface, the normal components of their gradients. The latter equivalent sources reflect the flow of current across the epicardial surface, and are thus suited for the imaging of regional ischemia and infarction. To study this formulation of the inverse problem, we used a realistically shaped boundary-element model of the human torso with an embedded model of ventricular activation. We then directly calculated epicardial and body-surface potentials for simulated activation wavefronts represented by oblique-double-layer sources, and used these test data for assessing possible advantages of the inverse solution in this amended formulation. Finally, we have shown how this inverse procedure might improve estimation of the extent and severity of regional ischemia.