Meeting Banner
Abstract #2770

Local SAR Control for Parallel Transmit MRI Using Multiple Patient Models

Hanno Homann1, Inagmar Graesslin1, Ulrich Katscher1, Peter Vernickel1, Kay Nehrke1, Peter Brnert1

1Philips Research Laboratories, Hamburg, Germany

Parallel transmission enables control of the RF transmit field in time and space. Hence, the B1-field and the SAR distribution can be influenced. To evaluate the potential for SAR reduction, numerical RF field simulations were carried out for an 8-channel Tx body coil and 9 volunteer body models.

Keywords

absorption according achieved additional applied applying arms available barrier body calculation calculations carried challenging channel clinical coefficient coil compressed compression compromises computation computational computationally consequence consider considered considering constrained constraints containing control converged convex cost decreased depend dependence dependency described design desirable dielectric discrimination distribution domain drive efficient effort eight elements enables entail every excitation excitations fast fatty field finite freedom function generalized generally hamburg hence highly homogeneity homogeneous human ignore important incorporated independent influenced inside instead laboratories least limits local magma magnitude maintaining margin marginal matrices matrix model models multitude necessarily nine normalized numerical often optimization parallel patient performance peter placed positions precomputed predict predicted prediction press probably projections quadrature real reasonable reduction relevant representing required respectively rich robust selected sensitivities separated separately setting severe shimming significantly simulated simulations simultaneous since slightly space spatially squares still substantial system tissues torso transmission transmit typically uncertainty variation vector volunteer volunteers water whole