RF Birdcage with ASTM Phantom and Generic Lead, 64 MHz

Model

The CAD birdcage model shown in the previous example is simulated with the ASTM phantom. 2 generic leads are placed in the phantom to study implant lead safety are localized SAR in the vicinity of the lead tips. The leads are modeled as PEC cylinders (diameter = 1.6 mm) with a dielectric insulation, which does not completely cover the tips of the leads. The model is part of  an interlab study, the results will be published at BEMS [1].

Simulation

In order to resolve the high field gradients at the tips of the leads a very high grid resolution must be used across the diameter (5 cells => min. grid step is 0.3 mm). The resulting FDTD time step is extremely small and causes very long simulation times even with hardware acceleration.

Simulation with Huygens Box

Using a Huygens Box excitation the total simulation time can be reduced by almost 10x from > 20 hrs to about 3 hrs. The process involves running 2 simulations:

simulation 1: the full birdgcage and ASTM phantom are simulated, but no leads. The near-field is recorded in 3D Field sensor in the around the location of the lead in the phantom.

simulation 2: the results of the 3D Field sensor from simulation 1 are used to feed the Huygens Box excitation. Only the lead and a small area of liquid needs to be simulated. The resulting grid only needs to cover a very smaller area and the number of cells is < 1 MCells. The minimum step is still small, but the simulation is much faster due to the small grid.

Results

Very high peak SAR distributions are observed at the exposed tips of the leads due to the E-field discontinuity. Using the Huygens Box excitation the simulation time can be signinficatly reduced.

 [1] Numerical and Experimental Evaluation of RF Induced Heating of a Generic Medical Implant During MRI Exams, E. Cabot et. al. BEMS 2008