Personalized Computational Framework to Study Arrhythmia Mechanisms on Top of ECG Image-Detected Substrate.

Electrocardiographic Imaging (ECGI) can unmask electrical abnormalities that were difficult to detect using the standard 12-lead ECG. However, it is still challenging to interpret the potential arrhythmogenic consequence of electrical patterns found with ECGI. Here, we introduce a computational framework that allows personalized simulations of cardiac electrophysiology (EP) to mimic electrical substrate as detected in an individual, to study the interaction between that substrate and premature ventricular complexes (PVCs). In patient data, electrical substrate identified using ECGI shows regions of pronounced dispersion of local recovery (i.e., recovery gradients). A computational model of ventricular EP was developed and then used to mimic the recovery gradients and PVCs found in patients. We studied a variety of gradients (6–98 ms/cm) and coupling intervals of the extra stimulus (−70 to +260 ms relative to the end of local recovery), which showed that re-entry can only occur when dispersion of recovery is large (≥76 ms/cm), and the extra stimulus occurs just after local recovery ended (~+40 ms). In conclusion, this computational framework allows to identify the specific conditions under which ECGI-detected substrates and PVCs can lead to re-entry in a personalized approach.