Brain scanning technologies: How does MEG stack up?

Magnetoencephalography (MEG) is a non-invasive imaging technology that has been used for decades in clinical and research settings. We're unlocking its value for more widespread adoption.

Magnetoencephalography (MEG) is one of several brain scanning technologies used to assess the brain. In today’s post, we look at MEG scanning and how it compares to several other types of scans, including MRI (magnetic resonance imaging), fMRI (functional magnetic resonance imaging), EEG (electroencephalography), and a few others. While none are as much of a mouthful as magnetoencephalography - they don’t exactly roll off the tongue either - so we will be sticking with acronyms from here on out.

There is no one singular “best” technology for looking at the brain. MEG and the other scanning modalities we are going to look at today all have their specific strengths, use cases, and limitations. 

Up first, MEG.

Click here to read our MEG primer for more of a deep dive analysis.

How MEG works: MEG passively and directly measures the magnetic fields produced by neurons in the brain as they fire. It passively records as a person’s brain works (i.e. as its neurons fire), akin to a voice recorder passively recording as a person talks.

MEG provides both high temporal resolution (accuracy with respect to timing of brain activity) and spatial resolution (accuracy with respect to where the activity is coming from). MEG’s ability to provide millisecond and millimeter accuracy of brain activity makes it valuable for both research and clinical applications. MEG scanners are large, expensive, and stationary scanners which have an appearance of an old fashioned 80’s hair dryer, and they typically sit in preeminent hospitals and research institutions (which is why to date, MEG has not typically been accessible to the general public). 

Where MEG is used: It is used extensively in research to study how brain activity changes in various states of health and disease, as well as when injured, and to understand cognitive processes. Clinically it’s used to localize epileptic activity, and for presurgical mapping like where in the brain functions like speech are located prior to neurosurgery.

MRI

How MRI works: MRI applies strong magnetic fields and radio waves to the body to create detailed images, or pictures, of how organs and tissues within the body look, thereby providing structural information.

Where MRI is used: Used for both clinical diagnosis (for detecting structural abnormalities like tumors, strokes, etc.) and research (for studying brain anatomy and development). Whereas the output of an MRI shows an individual an image of their anatomy, MEG is focused on neurophysiology, or the functioning of the brain. In this way, these two technologies serve different purposes, but they can complement each other and are often used together.

Think function for MEG, and structure for MRI. 

EEG

How EEG works: EEG passively and directly measures the electrical activity produced by neurons in the brain as they fire, using electrodes placed with gel on the scalp.

Where EEG is used: Widely used in clinical and research settings for diagnosing epilepsy, sleep disorders, and studying cognitive functions such as attention and memory. EEG provides the same type of information as MEG when it comes to measuring brain activity, but because EEG measures the electrical activity instead of the magnetic field that electrical activity produces, it is not as accurate as MEG in identifying exactly where in the brain that activity is coming from. However, it benefits from being highly portable and inexpensive in comparison to MEG.

fMRI

How fMRI works: This type of analysis measures changes in oxygenated blood flow related to neural activity: when a particular brain region becomes active, neurons in that area consume more oxygen and fMRI measures the amount of that oxygen. 

Where fMRI is used: Similar to MEG, fMRI is used extensively in research to study how brain activity changes in various states of health and disease, and to understand cognitive processes. Clinically it’s used for presurgical mapping. 

Blood flows at a slower speed than electrical activity fires, and therefore fMRI doesn’t have the same temporal accuracy that MEG has. However, nearly all fMRI scanners are also MRI scanners and therefore are far more accessible than MEG scanners. 

Other brain scanning technologies:

PET and SPECT: PET (Positron Emission Tomography) and SPECT (Single-Photon Emission Computed Tomography - seen above) are nuclear imaging scans that analyze brain function. Both scanning techniques involve injecting a small amount of radioactive material (tracer) into the bloodstream (more invasive than MEG or MRI). The scanners then detect abnormal cells that absorb large amounts of the radiotracer injected, which is indicative of a potential health problem.

fNIRS: Functional near-infrared spectroscopy (fNIRS) is a non-invasive neuroimaging technique that measures brain activity by detecting changes in blood oxygenation levels. Sound familiar? It should. fNIRS effectively measures the same thing as fMRI. However, fNIRS uses light absorption to measure blood oxygenation instead of applying a magnetic field and radio waves. Because it uses light, it predominantly measures brain activity closer to the skull as opposed to in the center of the brain, but it benefits from being a light and portable headset, whereas fMRI is a large stationary scanner. 

Why do we use MEG?

Unlike fMRI, PET, and SPECT, which measure indirect markers of brain activity (e.g., blood flow), MEG directly detects the magnetic fields produced by neuronal electrical currents. This provides a more direct representation of neural activity.

Besides its accuracy in directly measuring function, MEG is also completely non-invasive and passive. This means that you could theoretically be scanned as many times as you wanted, and for as long as you desired (annual brain scans as part of a regular preventive health routine, for example).

EEG is also a direct measure of brain function, but is harder to set up than MEG - it requires gel on the scalp and individual placement of electrodes - and is also less precise. This precision is particularly beneficial in both clinical diagnostics and cognitive neuroscience research.

Temporal resolution:

• MEG: Offers very high temporal resolution (in milliseconds), allowing researchers to track brain activity in real-time.

• EEG: Provides excellent temporal resolution, similar to MEG, because both directly measure electrical activity.

• fMRI: Relatively slow compared to MEG and EEG, with a temporal resolution in the range of seconds due to the sluggish nature of hemodynamic responses.

• MRI: Does not measure brain activity directly and thus does not have temporal resolution for studying fast neural processes.

Spatial resolution:

• MEG: Provides very high spatial resolution, sufficient to locate brain activity within a few millimeters.

• EEG: Offers low spatial resolution compared to MEG and MRI, as it detects signals from large areas of the brain rather than pinpointing specific locations.

• fMRI: Offers high spatial resolution, providing detailed anatomical information and functional localization within the brain.

• MRI: Provides the highest spatial resolution among these methods, offering detailed images of brain structures down to sub-millimeter scales.

Other differences with MEG:

• Tolerance for some movement: While not as motion sensitive as fMRI, MEG allows for more participant movement than traditional neuroimaging techniques, making it suitable for studying children or patients who have difficulty remaining still.

• Silent operation: MEG is silent during operation (unlike fMRI), making it more comfortable for participants and suitable for auditory studies.

• Complementary to other techniques: MEG data can be effectively combined with fMRI or EEG to leverage the strengths of multiple modalities.

Today’s post isn’t meant to be a sales pitch for MEG. We shared a few limitations to the technology, particularly related to accessibility and high equipment costs. To date, those barriers have been overcome only where MEG is most needed - applications requiring high temporal resolution and accurate source localization.

What we’re doing at MYndspan is expanding those applications to all people interested in early detection of neurodegeneration, and better tracking and measurement of their brain function and health over time.