Model-based dynamic off-resonance correction for improved accelerated fMRI in awake behaving nonhuman primates

Area of Science:

• Neuroimaging
• Biomedical Engineering
• Primate Research

Background:

• Accelerated functional MRI (fMRI) in nonhuman primates (NHPs) is crucial for neuroscience research.
• Vigorous body motion in awake NHPs introduces dynamic off-resonance, degrading image quality.
• Existing methods often require sequence modifications or extra acquisitions, limiting applicability.

Purpose of the Study:

• To estimate and attenuate dynamic off-resonance effects caused by body motion in accelerated NHP fMRI.
• To improve the robustness and reliability of fMRI data acquisition in awake, behaving NHPs.
• To reduce the gap between NHP and human fMRI imaging capabilities.

Main Methods:

• Utilized echo-planar imaging reference navigators to estimate k-space shifts caused by low-frequency, spatially linear off-resonance.
• Employed GRAPPA operators to compare navigator frames and quantify shifts.
• Applied estimated shifts for retrospective data correction on a frame-by-frame basis.

Main Results:

• Successfully estimated low-order dynamic off-resonance perturbations in phantom and simulation data.
• Demonstrated retrospective correction of corrupted data in simulations.
• Reduced ghosting artifacts and geometric distortions by up to 20% in simultaneous multislice in vivo NHP fMRI.

Conclusions:

• A novel, sequence-independent method for correcting dynamic off-resonance in accelerated NHP fMRI was developed.
• The method enhances imaging robustness without extra acquisitions, making NHP fMRI more reliable.
• This advancement facilitates more advanced NHP imaging protocols, comparable to human studies.