Bidomain modeling of cardiac tissue on a message-passing parallel computer

Due to the complexity of the cardiac tissue's dynamic properties, detailed multi-dimensional finite-difference bidomain modeling is computationally intensive and the model size is often limited by the computational power of a serial computer. We describe the use of a message-passing parallel computer to provide the memory and reduction in solution time for solving large finite-difference bidomain problems. Data distribution and inter-processor communication are achieved with the message-passing interface standard to ensure the portability of the computer code. Details on the parallel algorithms are presented together with some computational performance results.