Study on Search Strategies for Assessing the Worst Case RF-Induced Heating for Multi-Configuration Implant System Under MRI

This paper presents a study on strategies of searching for the worst case configuration of a multi-configuration implantable system under magnetic resonance imaging, which induces the highest radio frequency energy exposure among all possible configurations. The strategies are usually based on parameter search to avoid brutal study of a huge number of possible configurations. Comparison studies of four different searching strategies, i.e., single sequential search, sequential search with the second worst case validations, hotspots-based sequential search, and multi-iteration sequential search, are evaluated. It is found that the initial configuration and sequence of parameter search can significantly affect the searched worst case configuration. For the multi-configuration system considered in this investigation, the simple sequential search strategy leads to a high probability of search failure, which means that the worst case configuration may be easily missed. Single sequential search with the second worst case validations and multi-iteration sequential search can significantly reduce the probability of search failure.