Local SAR management by RF shimming: a simulation study with multiple human body models.

Area of Science:

• Magnetic Resonance Imaging (MRI)
• Biomedical Engineering
• Computational Electromagnetics

Background:

• Parallel transmission offers direct control of radiofrequency (RF) fields in MRI.
• It can improve RF field homogeneity and potentially reduce specific absorption rate (SAR), enabling advanced sequence design.
• Predicting local SAR is complex due to multi-channel drive and patient-specific anatomy.

Purpose of the Study:

• To investigate RF shimming for SAR management in high-field MRI.
• To evaluate the impact of patient-specific models on SAR prediction.
• To develop and assess a novel RF shimming approach for enforcing local SAR constraints.

Main Methods:

• Finite-Difference Time-Domain (FDTD) simulations were performed using a 3 T body coil with eight transmit elements.
• Nine human body models derived from volunteer MR data were utilized to account for patient variability.
• A new RF shimming method was proposed to enforce local SAR limitations.

Main Results:

• RF shimming significantly reduced local SAR across all simulated volunteers.
• Implementing SAR constraints led to further SAR reduction with minimal impact on RF performance.
• The proposed method demonstrated consistent SAR reduction for various body models.

Conclusions:

• Parallel transmission is a valuable tool for controlling and managing local SAR in the human body.
• The practical application of local SAR constraints is feasible and yields consistent results.
• This approach enhances safety and flexibility in high-field MRI sequence development.