Robust selective weighted field mapping using multi-echo gradient echo-based MRI.

Area of Science:

• Medical Imaging
• Biophysics
• Magnetic Resonance Imaging

Background:

• 3D gradient echo (GRE) and echo planar imaging (EPI) in MRI suffer from signal loss and image distortion at air/tissue interfaces due to macroscopic field gradients.
• Conventional field mapping techniques often ignore magnitude data and struggle with noise and phase aliasing, especially at longer echo times (TE) and echo spacings (ΔTE).

Purpose of the Study:

• To develop and validate a novel, noise-resilient method for computing accurate low and high-resolution magnetic field maps across all field map regimes (small and large echo spacing).
• To improve the robustness of field mapping in challenging low signal-to-noise ratio (SNR) regions common in clinical MRI.

Main Methods:

• An analytical approximation and a novel computational method were developed, weighting the field offset by magnitude signal decay quality to minimize noise dependence.
• The method accommodates various echo time (TE) and echo spacing (ΔTE) regimes and complex phase wrap scenarios.
• Performance was evaluated using simulated data, experimental phantoms, and in vivo human studies.

Main Results:

• The proposed method significantly outperforms conventional techniques, especially in low SNR regions (e.g., near the sinus cavity, brain edges) with mean errors less than 1 Hz.
• In high SNR regions (SNR > 20), it achieves comparable accuracy to conventional methods (mean error ~0.1 Hz).
• The technique demonstrates enhanced robustness against noise and phase aliasing.

Conclusions:

• The novel field mapping technique offers improved accuracy and robustness across diverse MRI acquisition parameters and SNR levels.
• It shows significant promise for enhancing clinical MRI by providing better field map generation, particularly in challenging anatomical areas.
• The method is easily implementable for rapid computation in clinical settings.