Computational Biology of cardiac arrhythmias: from basic science to application

Cardiac virtual tissues are biophysically, histologically and anatomically detailed computational models that are sufficiently well validated to be used as a predictive tool, are currently used in basic research, and are beginning to be applied to clinical problems. Virtual cardiac cells and tissues are stiff, high order ordinary and partial differential equations. While 1and 2-D tissues can be run on a single CPU, 3-D tissues are more suitable for SMP parallel computation. Human virtual cardiac tissues have been developed, and clinical application to individual patients requires faster patient specific reconstruction from multi-modal clinical data sets. Implementation of this patient specific approach will require high bandwidth access from tertiary clinical centres to teraflop compute resources. In principal this could be met by Grid technologies, in practice dedicated HPC clusters may be required.