Optimization of MRI Gradient Coils With Explicit Peripheral Nerve Stimulation Constraints.

Area of Science:

• Medical Imaging
• Biophysics
• Electrical Engineering

Background:

• Peripheral Nerve Stimulation (PNS) is a significant limitation in Magnetic Resonance Imaging (MRI), particularly for fast imaging sequences that utilize strong gradient systems.
• Current methods assess PNS characteristics post-coil design, hindering efficient design modifications to mitigate this issue.
• This retrospective assessment makes it challenging to proactively identify and implement design changes that could reduce PNS.

Purpose of the Study:

• To introduce a direct computational approach for integrating Peripheral Nerve Stimulation (PNS) effects into the MRI coil optimization process.
• To develop a framework enabling the identification of coil designs that minimize PNS while adhering to standard engineering requirements.
• To explore the potential for enhancing MRI acquisition rates and safety through optimized coil design.

Main Methods:

• Incorporation of knowledge regarding magnetic field interactions with peripheral nerves into an optimization algorithm.
• Development of a simulation framework to directly assess and minimize PNS during the coil design phase.
• Comparison of simulated PNS thresholds for PNS-optimized gradient coils against conventional designs.

Main Results:

• Demonstrated a direct method to incorporate PNS effects into the coil optimization process.
• Identified coil solutions that minimize PNS while meeting traditional engineering constraints.
• Achieved up to a 2-fold reduction in PNS propensity in simulated head and body gradients compared to conventional designs.
• Observed moderate trade-offs in coil inductance and field linearity.

Conclusions:

• The developed framework allows for the direct optimization of MRI coil designs to minimize Peripheral Nerve Stimulation (PNS).
• PNS-optimized coils show a significant reduction in PNS propensity, potentially enabling faster and safer MRI acquisition.
• This approach holds promise for improving imaging performance and could be extended to designing stimulation devices.