Computational simulation and analysis of 3D body surface potential patterns generated by common atrial arrhythmias

Atrial arrhythmias are the most common cardiac pathologies and the variety of mechanisms of onset and progression are not well understood. This study proposes a set of indices able to characterize a normal sinus rhythm and discriminate between flutter, tachycardia and fibrillation. Attention was paid to the correlation between atrial sites with high frequency arrhythmic activity and the corresponding torso surface activity to help in the non-invasive diagnosis of electrical cardiac problems. A 3D human torso model that includes an anatomically realistic atria model was used to obtain biophysical simulations of electrical atrial depolarization and surface potential maps. Dominant frequency, power spectral peak, spectral correlation, sample entropy and phasespace diagrams were studied and graphically represented by 3D maps to describe characteristic patterns. These indices and gradients in surface maps revealed important dissimilarities in spatial and temporal distributions among dif erent propagation patterns and allowed classifYing each type of arrhythmia.