Self-organizing network simulation of cardiac contraction dynamics.

Self-organizing network encodes and resembles structural geometry in the heart, offering a new pathway to study cardiac simulation. Hence, we have leveraged the sparsity of an adjacency matrix to design novel simulations of cardiac electrical dynamics on the self-organizing network. However, very little has been done to investigate network simulation of mechanical contraction dynamics. As a vertical step, this paper presents a new self-organizing network methodology for simulation modeling of cardiac contraction dynamics. To this end, we propose to model the self-organizing network as an interconnected spring-mass-damper system and further solve networked dynamic equations to simulate the orchestrated dynamics of mechanical contractions. The proposed methodology is evaluated and illustrated on both 2D cardiac tissues and the 3D heart. Experimental results show that the proposed methodology not only effectively models contraction dynamics in excitable media, but can also be flexibly extended to the whole heart.