Shape Optimization of an Electric Dipole Array for 7 Tesla Neuroimaging.

Area of Science:

• Magnetic Resonance Imaging (MRI)
• Radio-Frequency (RF) Engineering
• Biomedical Engineering

Background:

• Ultra-high field (UHF) MRI at 7 Tesla (T) offers enhanced signal-to-noise ratio.
• Radio-frequency (RF) arrays using electric dipoles show promise for UHF MRI transmit and receive applications.
• Implementation challenges exist for dipole RF arrays in human head imaging at 7 T.

Purpose of the Study:

• To examine implementation barriers of dipole RF arrays for 7 T human head MRI.
• To design and optimize a conformal, meandered dipole RF array for improved performance.
• To evaluate the transmit and receive performance of the optimized dipole array.

Main Methods:

• Dipole array construction with conformal, meandered dipoles.
• Evolutionary-based optimization routine for dipole dimension selection and shape optimization.
• Coupling matrix synthesis (CMS) for dipole decoupling.
• Evaluation of transmit efficiency, specific absorption rate (SAR), and noise correlations.
• Assessment of parallel imaging performance as a receiver.

Main Results:

• Achieved mean and worst-case nearest-neighbor dipole transmission of -17.2 and -15.5 dB, respectively.
• Observed transmit efficiencies of 24.6 nT/V (whole brain) and 26.0 nT/V (axial slice).
• Reported total and peak 10-g SAR of 0.163 and 0.601 W/kg (normalized to 1 W input power per channel).
• Demonstrated high element orthogonality with maximum and mean noise correlations of -17 dB and -32 dB.
• Optimized array showed improved transmit uniformity and reduced 10-g SAR compared to non-optimized designs.

Conclusions:

• The combination of CMS and shape optimization effectively designed a dipole array with sufficient transmit uniformity.
• The optimized dipole array achieved a reduction in 10-g SAR compared to a non-optimized array of the same geometry.
• The dipole array maintained high receiver element orthogonality, enabling strong parallel imaging performance.