RESUMO
PURPOSE: Many procedures involving catheters and implanted medical devices could benefit from MRI guidance but are currently contraindicated due to risk of significant heating near linear conductive structures. A priori safety prediction is impossible in vivo and thus, safety is typically investigated in vitro by directly measuring temperature rise. Existing methods of investigating safety are inflexible and provide few data. Furthermore, they are fundamentally limited because dangerous temperatures rises can only be investigated if induced. A method of remotely predicting safety is necessary for ensuring safety in patients. THEORY AND METHODS: Electric current induced on the metallic object causes any dangerous heating; thus a remote method of safely characterizing the induced radiofrequency (RF) current distribution would suffice to evaluate safety assuming conservative estimates for local tissue properties. Here we propose a method of analyzing induced phase artifacts seen in low-specific absorption rate characterization images, to determine induced current on an interventional device. This induced current distribution can then be used to predict RF heating behavior under application of any other imaging sequence. RESULTS: This method has been successfully used to reproduce numerical simulations in a phantom. Furthermore, the heating behavior around a conductive wire produced by a scan other than that used to characterize current was successfully predicted. CONCLUSION: It has been shown in phantom experiments that remote current characterization can safely prevent dangerous scans as well as enable safe scans that previously would not have been attempted.