Cardiac myocyte model parameter sensitivity analysis and model transformation using a genetic algorithm

Cardiac arrhythmia is the disruption of the normal electrical rhythm of the heart and is a leading cause of mortality around the world. To study arrhythmogenesis, mathematical models of cardiac myocytes and tissues have been effectively employed to investigate cardiac electrodynamics. However, among individual myocytes, there is phenotypic variability that is dependent on factors such as source location in the heart, genetic variation, and even different experimental protocols. Thus, established cardiac myocyte models constrained by experimental data often are untuned to new phenomena under investigation. In this study, we show direct links to parameter changes and differing electrical phenotypes. First, we present results exploring model sensitivity to physiological parameters underpinning electrical activity. Second, we outline a genetic algorithm-based approach for tuning model parameters to fit cardiac myocyte behavior. Third, we use a genetic algorithm to transform one model type to another, relating simulation to experimental data. This model transformation demonstrates the potential of genetic algorithms to extend the utility of cardiac myocyte models by comparing different functional regions in the heart.