Robust RF shimming and small-tip-angle multispoke pulse design with finite-difference regularization | JoVE Visualize

Area of Science:

Magnetic Resonance Imaging (MRI)
Biomedical Engineering
Signal Processing

Background:

Ultrahigh-field MRI (7 Tesla and above) presents challenges in radiofrequency (RF) field homogeneity.
Parallel transmit RF shimming and small-tip-angle multispoke pulse design are crucial for optimizing image quality and minimizing artifacts.
Existing magnitude least-squares algorithms can suffer from flip angle nulls, leading to signal loss and reduced robustness.

Purpose of the Study:

To introduce and evaluate a novel finite-difference regularizer for magnitude least-squares optimization.
To improve the robustness of parallel transmit RF shimming and small-tip-angle multispoke pulse designs in ultrahigh-field MRI.
To mitigate flip angle nulls and enhance image signal-to-noise ratio (SNR).

Main Methods:

A finite-difference regularization term was integrated into an iterative magnitude-least-squares pulse design algorithm.
The regularizer was activated upon detection of unwanted flip angle null distributions.
The algorithm's performance was assessed using simulated and experimental Bloch simulation data ( maps) at 7 T across various transmit arrays and human subjects.
Dynamic multislice RF shimming for single-shot gradient-echo EPI in human functional MRI at 7 T was used for experimental validation.

Main Results:

The finite-difference regularizer effectively prevented the formation of excitation nulls in RF shimming and small-tip-angle multispoke pulses.
A monotonic trade-off relationship between flip angle error and RF power was observed, improving pulse design.
Simulations and experiments with multiple head-array geometries and 12 healthy subjects confirmed the algorithm's effectiveness.
Functional MRI scans at 7 T with dynamic RF shimming demonstrated high image SNR throughout the brain, contrasting with significant signal loss using conventional methods.

Conclusions:

Finite-difference regularization significantly enhances the robustness of RF shimming and small-tip-angle multispoke pulse design algorithms.
The proposed method leads to improved flip angle homogeneity and a predictable relationship between flip angle error and RF power.
This approach effectively avoids unwanted solutions, paving the way for more reliable ultrahigh-field MRI applications.