Modeling of cardiac stimulation limits by MRI gradient coils in a population of body models

Motivation: Our previous modeling of gradient-induced cardiac stimulation (CS) in two body models indicated that the regulatory IEC 60601-2-33 CS limit overestimated CS thresholds by 9- to 46-fold. Goal(s): To investigate the expected variance of CS thresholds across a healthy population. Approach: We deploy our validated cardiac magnetostimulation modeling approach in six body models with varying shape/BMI/age and a commercial gradient system. Results: Predicted CS thresholds vary up to 2-fold across body models. Worst-case CS thresholds are 7X greater than the IEC CS limit and 4X greater than experimental PNS limits. Impact: Our modeling allows investigation of the variability of CS thresholds across the population, which is not accessible experimentally. This knowledge is critical to obtain a robust estimate of safe, but not overly restrictive cardiac safety limits for MRI gradients.