Session SA3.1
Comparison of Microscopic and Bidomain Models of Anisotropic Conduction
JG Stinstra*, CS Henriquez, RS MacLeod
University of Utah
Salt Lake City, UT, USA
One of the most popular modeling approaches to simulate the electrical potentials within the myocardium is the so-called bidomain model. This model reduces the discrete structure of the myocardium (individual myocytes coupled by gap junctions) by a set of continuous intracellular and interstitial domains coupled by an active membrane. In the bidomain model, the electrical properties of tissue are represented by two continuously varying tensors that denote the apparent conductivity of the intracellular and extracellular spaces, respectively. We wished to investigate the relationship between these bidomain conductivity tensors and the discrete features of cardiac tissue under both normal conditions and during pathophysiological episodes. One way to carry out this analysis is to generate a detailed model of cardiac tissue that incorporates histological details such as extracellular space and gap-junction distribution as well as myocyte shape. In this detailed model, one can simulate the propagation of a planar wavefront along and across the fiber direction of a small patch of tissue and compute from this the apparent resistances of the intracellular and extracellular spaces. Our specific hypothesis was that a bidomain model using these apparent resistances would produce the same conduction velocities and depolarization wave profiles as a more detailed discrete model, but with greater efficiency. To test this hypothesis we generated a histological model of cardiac tissue that represented a healthy state, modeled after observations from histology and confocal images of intact myocardium. The simulation study results showed that under normal healthy conditions, propagation velocities along and across the fiber matched those computed with the bidomain very well (less than 5% difference). However, under ischemic conditions, in which the amount of extracellular space was reduced by a factor 2 and the gap junctional resistance increased by a factor 5, the apparent resistances could no longer be used as good estimates in the bidomain model for waves traveling across the fiber direction; differences in conduction velocities between the two models were larger than 20%. In conclusion, under healthy conditions the bidomain tissue properties can be approximated by the macroscopic apparent resistances, however when the pathways within the tissue become more resistive, additional modeling is needed to estimate accurate parameters.
(Abstract Control Number: 118)