Cross-term-compensated gradient waveform design for tensor-valued diffusion MRI.

Area of Science:

• Magnetic Resonance Imaging
• Biophysics
• Medical Physics

Background:

• Diffusion MRI is sensitive to background magnetic field gradients, leading to signal errors.
• These errors, particularly 'cross-terms,' compromise the accuracy of diffusion MRI measurements.
• Accurate parameterization in diffusion MRI is crucial for understanding tissue microstructure.

Purpose of the Study:

• To develop a novel gradient waveform design for diffusion MRI that cancels background gradient-induced cross-terms.
• To improve the accuracy and robustness of diffusion MRI measurements in the presence of static background gradients.
• To provide a versatile optimization framework for advanced diffusion MRI sequence design.

Main Methods:

• Numerical optimization was used to design gradient waveforms, maximizing encoding efficiency while constraining cross-term sensitivity (c=0).
• The proposed waveforms were tested against previous designs for various b-tensor encoding types (linear, planar, spherical).
• Experimental validation was performed on a clinical MRI system using a water phantom with known background gradients.

Main Results:

• Optimized cross-term-compensated waveforms demonstrated superior performance compared to previous designs across different encoding types.
• Experimental results showed the compensated waveforms were robust to background gradients, unlike conventional designs.
• Simulations confirmed robustness to background gradients (0-3 mT/m) and reduced signal/parameter bias compared to non-compensated designs.

Conclusions:

• A novel, efficient cross-term-compensated gradient waveform design for diffusion MRI has been developed and validated.
• This design facilitates accurate diffusion MRI measurements, robust to static background gradients of varying amplitude and direction.
• The open-source optimization framework supports diverse sequence parameters and advanced encoding techniques.