A Computational Method for the Analysis of Fast and Transient Regimes that Determine Arrhythmia Formation and Termination

Aaron Gobeyn1, Desmond Albert Kabus2, Elena G. Tolkacheva3, Hans Dierckx1
1KU Leuven, 2KU Leuven & LUMC, 3University of Minnesota


Abstract

Aims: The precise mechanisms that cause and sustain complex arrhythmias are still under investigation. Various computational tools have been proposed to assist clinicians in their arrhythmia treatment therapies. Several approaches have been developed to specifically target rotors as one of the arrhythmias mechanisms, aiming to identify phase singularities. Other techniques search for reentrant circuits, conduction block lines and focal sources. During the crucial processes of arrhythmia initiation and termination, it is imperative to determine whether conduction blocks will grow thus causing an arrhythmia, or shrink thus simplifying the pattern's complexity.

Methods: Recently, we demonstrated that conduction block lines can be regarded as extended regions of nearly discontinuous phase, or phase defects. This observation implies that at the center of a linear-core rotor, the wave front and wave back can end in different points, which we termed as 'heads' and 'tails', respectively. Here, we propose a fast computational method to locate the wave fronts, wave backs, heads and tails in spatiotemporal maps of cardiac excitation. Via bitwise operations, these points can be labeled in a phase map nearly instantaneously.

Results: The method was applied to simulations and optical voltage mapping data of ventricular tachycardia in pigs. We find that in steady regimes, these points of interest tend to co-localize, conforming to the classical phase singularity view. However, the formation of a figure-of-eight reentry (double spiral) is seen to involve different steps where the heads and tails form intermediate bound states that have not been reported before. In non-periodic tachycardia recordings, our detected points of interest were more robust than classical phase singularities.

Conclusion: We designed and implemented a computational framework that refines the classical phase singularity detection. Our method can capture short-lived and smaller-scale structures than before. The algorithm will be made available to the public to help elucidating arrhythmia mechanisms.