Universal pulses: A new concept for calibration-free parallel transmission.

Area of Science:

• Magnetic Resonance Imaging
• Biophysics
• Medical Physics

Background:

• Radiofrequency (RF) field inhomogeneity is a significant challenge in high-field MRI, particularly at 7 Tesla (T), leading to image artifacts and reduced diagnostic accuracy.
• Parallel transmission (Tx) techniques offer a solution by using multiple RF coils to shape the RF field, but often require subject-specific calibration and optimization.

Purpose of the Study:

• To investigate a calibration-free parallel transmission method for mitigating RF field inhomogeneity in 7T brain imaging.
• To develop and validate "universal" RF pulses that are robust across a population, eliminating the need for individual field mapping and pulse optimization.

Main Methods:

• A database of RF and static field maps was created from six volunteers at 7T.
• Small-tip-angle and inversion pulses were designed using joint k_T-points trajectory optimization for population robustness.
• These "universal" pulses were implemented in an MPRAGE sequence on six additional volunteers, with comparisons to circularly polarized and subject-optimized modes.

Main Results:

• Universal pulses achieved a normalized root mean square error (NRMSE) of approximately 11% for excitation and inversion.
• This significantly outperformed circularly polarized (NRMSE ~28%) and RF shim modes (NRMSE ~20%) across all subjects.
• Subject-based optimized pulses yielded slightly better results (NRMSE ~7%), but universal pulses demonstrated robust performance without individual tuning.

Conclusions:

• RF pulses can be designed to reliably address RF field inhomogeneity in a population, representing a step towards "plug and play" parallel transmission.
• This approach allows for offline pulse design, removing the necessity for subject-specific field map measurements.
• The findings pave the way for more accessible and efficient high-field MRI protocols.