Predictive models and simulations of cardiac function require accurate representations of anatomy, often to the scale of local myocardial fiber structure. However, acquiring this information in a patient-specific manner is challenging. Moreover, the impact of physiological variability in fiber orientation on simulations of cardiac activation is poorly understood.
To explore these effects, we implemented biventricular activation simulations using rule-based fiber algorithms and robust uncertainty quantification techniques to generate detailed maps of model variability. Specifically, we utilized polynomial chaos expansion, enabling efficient exploration with reduced computational demand through an emulator function approximating the underlying forward model. Our study focused on examining the epicardial activation sequences of the heart in response to six stimuli locations and five activation feature metrics.
Our findings revealed that physiological variability in fiber orientation does not significantly affect the location of activation features, but it does impact the overall spread of activation. For all stimuli, two trends emerged: (1) epicardial stimuli generate activation sequences with higher variability than endocardial stimuli, and (2) endocardial stimuli produce high variability across the rest of the epicardial surface. Furthermore, we observed low variability for the earliest and latest activation sites but high variability for the orientation and area of the breakthrough site.
The variation we observed may have minimal impact on clinical procedures that localize focal arrhythmias, which rely primarily on identifying sites of early activation. On the other hand, we did observe a more considerable degree of variability concerning the entire activation sequence. Such variability may impact modeling re-entrant arrhythmias as changes in the activation sequence based on fiber orientation may cause differences in the re-entry site. We conclude that the level of accuracy of myocardial fiber orientation required for simulation depends highly on the specific goals of the model and the related research or clinical goals.