Skip to Content

Demo: Magnetic Brain Imaging

William Sutherling of the Huntington Medical Research Institute demonstrates how to use magnetic imaging to hunt down seizure-causing brain tissue.

An epileptic seizure is the outward sign of an electrical storm in the brain, a sudden surge of uncontrolled electric currents. If neurosurgeons can pinpoint the damaged brain tissue that sparks the storm, they can remove it, potentially sparing a patient a lifetime of debilitating attacks and antiseizure medications. But zeroing in on the precise bits of defective gray matter using the scalp electrodes of a standard electroencephalograph (EEG) machine is difficult, because electrical fields generated in the brain “get spread out and distorted” as they pass through the skull, says William Sutherling, a neuroimaging expert at the nonprofit Huntington Medical Research Institutes (HMRI) in Pasadena, CA. So Sutherling looks as well to the magnetic fields generated by each electrical impulse in the brain; those pass through the skull virtually unaffected. Using one of only a few dozen magnetoencephalography (MEG) machines in the world, Sutherling is measuring the vanishingly faint magnetic fluctuations generated by epilepsy sufferers’ brains, and combining that data with 3-D information from magnetic resonance imaging (MRI). His hope is to prove that the method is a reliable, practical way to narrow and delimit the sources of seizures, so that surgeons can remove the offending tissue without damaging the healthy, functioning cells around it. This fall, Sutherling gave TR senior editor Wade Roush a tour of HMRI’s $2.5 million MEG facility and demonstrated how his team gathers data from the hopeful patients who venture into his chamber.

1. A Room with No View. Sutherling strides into the MEG chamber, a magnetically shielded room-within-a-room with an interior floorspace of about 10 square meters. The room houses HMRI’s whole-head MEG unit – so named because its array of internal sensors fits snugly over a patient’s head like a giant hair dryer. The sensors detect even the tiniest changes in any magnetic fields threading through them. Such fields are normally all around us, so tracing fluctuations to specific areas of the brain with millimeter-scale accuracy would be impossible unless the sea of ambient magnetic waves generated by fluorescent lights, computers, power lines, and the earth itself – not to mention the nearby MRI machine, which is essentially a giant magnet – were kept out.

2. Magic Metal. The secret to shielding the MEG unit, Sutherling explains, is a thin layer of an alloy called “mu metal,” visible between the blue exterior pane and the layer of white foam on the edge of the room’s bank-vault-like door. Once the door is closed, exterior magnetic fields flow around the mu metal cage, leaving the interior magnetically silent.

3-4. Head Cold. Superconductivity is the key to the MEG unit’s exquisite sensitivity. At the core of HMRI’s unit, built by VSM MedTech of Coquitlam, British Columbia, is an array of small metal rings called superconducting quantum interference devices, or SQUIDs, which look much like the suckers of an actual squid’s tentacles (3). When a ring is cooled to temperatures just above absolute zero, it becomes a superconductor, meaning that an electrical current traveling around it encounters virtually no resistance and could, in principle, keep circling forever. That current, in turn, produces a magnetic field. “If a magnetic field spreads through from the brain, it opposes the magnetic field already in that ring,” says Sutherling. “And the ring reacts by trying to keep the total current going through that ring the same.” Any new current induced in the ring causes a change in voltage that can be amplified thousands of times over and precisely measured by electronics. To stay superconducting, the SQUIDs must reside inside a huge flask of liquid helium, Sutherling says, touching the ungainly – and frosty – apparatus that fits around the patient’s head (4).

5. Another Reason to Brush. Before a patient is seated in the MEG unit, every piece of metal on his or her body must be demagnetized. Otherwise, the merest jiggle would break the magnetic calm. MEG technician Nancy Lopez has a handheld demagnetizer for this purpose; it’s powerful enough to neutralize even a patient’s dental fillings.

6. Sound Check. The first step in a MEG exam is to plumb the patient’s brain for fixed reference points, needed later to align the MEG data with MRI images and reckon the locations of electrical disturbances. Lopez outfits a subject with headphones so that she can administer a series of tones, which cause the brain’s auditory centers to sprout magnetic fields. These centers – which are well-known landmarks in the brain, always located on specific folds in the right and left temporal lobes – show up clearly on the MEG unit’s computer readout.

7. Electric Dreams. For the next hour or two, the patient must sit absolutely still – it’s okay to doze off – while the detector array observes the brain’s spontaneous electromagnetic activity.

8. Number Cruncher. Each of the SQUIDs sends its readings to a separate board in the MEG unit’s main computer, a refrigerator-sized behemoth across the room from the MEG chamber.

9. Skull Cap. At a bank of desktop and laptop computers adjacent to the main computer, Lopez and Sutherling open windows depicting the MEG unit’s measurements graphically. A map caricaturing a top view of the subject’s head, complete with a tiny triangular nose and elephant ears, shows the detectors’ locations around the skull as red circles. In another window, the changing readings from each individual detector are expressed as squiggly, EEG-like lines. The relatively flat readings from this healthy volunteer indicate that she’s asleep.

10. A Passing Storm. Conditions are quite different during a seizure. To illustrate, Sutherling calls up the records of an actual epilepsy patient examined in HMRI’s MEG unit before surgery. Rather than wait for a patient to have a spontaneous seizure during the exam, doctors implant a grid of electrodes just inside the skull, over the region of the brain thought to be affected. A jolt from these electrodes induces a miniseizure whose magnetic signature can then be recorded in detail. The individual SQUID readings from these small seizures are translated by software into a schematic showing where the strongest magnetic fields emanate from the skull.

Pointing to the three views of the head at the top of the screen, Sutherling explains that the magnetic readings are mathematically transformed into three dimensions and overlaid on MRI images of the patient’s brain. The stark yellow markings then guide surgeons to the wellsprings of a patient’s epileptic seizures – usually tiny bits of scar tissue.

Sutherling recalls one lifelong epilepsy sufferer whose seizures struck every two hours. EEGs showed unusual activity across the man’s frontal lobes, but MEG images traced the problem to a single spot in the left frontal lobe, near the speech center. Surgeons excised most of the scarred tissue while avoiding cuts that might have affected the man’s ability to speak. After the operation, he experienced only minor seizures. “Ideally, we want to make certain that the area that’s removed has zero function – that it’s just scar tissue – and that the removal is complete,” says Sutherling. “The goal is to make people seizure-free, so that they’re able to drive and able to work.”

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.