Exploring the Sensitivity Limits of Neuronal Current Imaging With MRI and MEG in the Human Brain

Area of Science:

• Neuroimaging
• Magnetic Resonance Imaging (MRI)
• Electrophysiology

Background:

• Conventional BOLD-fMRI indirectly measures neural activity via hemodynamic responses.
• Developing direct measures of neuroelectrical activity is crucial for brain imaging.
• Spin-lock (SL) fMRI shows potential for detecting neuronal activity but in-vivo feasibility is uncertain.

Purpose of the Study:

• To evaluate the in-vivo sensitivity and localization capabilities of SL-fMRI for detecting human neuronal activation.
• To benchmark SL-fMRI against magnetoencephalography (MEG) and BOLD-fMRI.
• To assess the feasibility of MR-based neuronal current imaging.

Main Methods:

• Thirteen healthy volunteers underwent 8 Hz visual stimulation.
• SL-fMRI (using rotary excitation and stimulus-induced rotary saturation) was performed alongside BOLD-fMRI and MEG.
• Postprocessing involved regression-filtering-rectification for stimulus-locked signal fluctuations.

Main Results:

• MEG detected robust stimulus-locked responses (~0.07 nT) in the occipital cortex.
• BOLD-fMRI confirmed reliable hemodynamic activation.
• Neither REX nor SIRS SL-fMRI methods produced consistent in-vivo stimulus-related activation.
• Phantom experiments showed detection thresholds (0.2–0.6 nT) exceeding physiological field amplitudes.

Conclusions:

• Current SL-fMRI implementations at 3T lack sufficient sensitivity for reliable in-vivo detection of neuronal magnetic fields.
• Physiological neuronal magnetic field amplitudes in the visual cortex are below the detection limits of the tested SL-fMRI techniques.
• Findings provide quantitative constraints for direct neuronal current imaging with MRI and guide future methodological advancements.