Search

This study investigates the radio frequency (RF)-induced heating for a common medical implanted device, the trauma screw, in different human models under 1.5 T and 3 T magnetic resonance imaging (MRI) systems. Three different human models, Duke, Ella, and Fats from the virtual family project, are used in the simulations. The human models and the device are loaded in a generic 32 ports RF coil. The screws with suitable sizes are implanted in four regions (hip, knee, shoulder, and wrist) for each human model. From the simulation results, the RF-induced heating near the device shows significantly difference within the models. Consequently, the clinically relevant RF induced heating can be patient specific and needs to be investigated individually.

This paper presents a comparison study of searching strategies for the worst-case construct for an implantable internal fixation system under 1.5T MRI radio frequency (RF) birdcage coil. An example of the generic plate and screw system is used as the device of interest. Based on the combinations of the values of the six device parameters, 576 models are studied and the peak 1 g averaged specific absorption rate (SAR) near devices are extracted to build a dataset for searching. Three searching strategies, which are sequential one-dimension searching, searching based on various sequences and searching with multi-iterations, are investigated. It is found that the searching sequences and initial values of the device parameters have effects on the searched worst-case results. The possibility of getting the failed search results when searching with random sequence and initial parameter values can be high. Searching with multi-iterations can improve the accuracy of correctly identifying the worst-case heating construct.