Transmural imaging of ventricular action potentials and post-infarct substrates in swine hearts

Transmural electrophysiological imaging (TEPI) using surface measurements is an ill-posed problem that does not have a unique solution in its general unconstrained form. We previously developed and preliminarily validated a Bayesian approach to TEPI that incorporates physiological, spatiotemporal priors through probabilistic integration of dynamic electrophysiological models. The main contribution of this study is to generalize this method of TEPI, which previously considers data from body-surface electrocardiograms (ECG), to use inputs from electroanatomical voltage mapping (EAVM) on heart surfaces. Feasibility of this new application of TEPI is positively verified on two swine hearts. Furthermore, we involve multiple reference data in forms of in-vivo EAVM and ex-vivo magnetic resonance imaging (MRI) of the post-infarct scar, allowing a detailed comparison of the three different modalities (EAVM, MRI & TEPI) for imaging/mapping post-infarct substrates. On the two swine models shared by STACOM 2011 EP simulation challenge, we are able to preliminarily demonstrate that, compared to the low-voltage surface scar area detected by EAVM, the 3D scar volume detected by TEPI is more consistent with scar volume delineated in MR images.