Properties of MEG tomographic maps obtained with spatial filtering.

Area of Science:

• Neuroscience
• Biophysics
• Medical Imaging

Background:

• Magnetoencephalography (MEG) is a leading noninvasive technique for studying brain oscillations.
• Spatial filtering methods enable whole-brain functional tomographic maps of coherence and power.
• MEG sensor sensitivity profiles lead to anisotropic spatial resolution across the brain.

Purpose of the Study:

• To develop a method for calculating the spatial resolution of MEG functional maps at any brain location.
• To analyze how signal-to-noise ratio and leadfield variations affect spatial resolution.
• To introduce a confidence volume for localizing activation maxima with quantified uncertainty.

Main Methods:

• Derived an analytic expression for computing spatial resolution across the entire brain.
• Investigated the relationship between resolution, signal-to-noise ratio, and leadfield changes.
• Developed a bootstrap-based method to compute a confidence volume for localization uncertainty.

Main Results:

• An analytic expression for spatial resolution computation was successfully derived.
• Resolution was shown to depend on signal-to-noise ratio and leadfield characteristics.
• A confidence volume provides a measure of localization uncertainty for MEG data.

Conclusions:

• The developed method provides a convenient way to compute and visualize spatial resolution maps for MEG.
• Resolution maps can be overlaid on anatomical MRI, similar to functional maps.
• The confidence volume aids in precise anatomical localization of brain activity and statistical comparisons.