This Device Reads Paralyzed People’s Thoughts Through Their Veins

A brain implant inserted in the jugular allows ALS patients to type with their minds

Photos: Synchron

Graham Felstead first noticed his left arm becoming weak in 2016. Two years later, he was diagnosed with amyotrophic lateral sclerosis, or ALS, a neurological disorder that culminates in paralysis of most voluntary muscles. Eventually, he lost the ability to use his arms and became completely reliant on his wife for the most basic of everyday tasks.

But after receiving a brain implant in August 2019, he’s gained back some of the independence. The implant, which connects wirelessly to his personal computer and other smart devices, allows Felstead, 75, to surf the web, check his email, and write Word documents using just his thoughts. Together, the system is known as a brain-computer interface.

“The device has allowed me to be productive again, including shopping, banking, and delegating tasks among the Rotary Club members with whom I volunteer,” he said in a statement released by Synchron, the Silicon Valley-based company that made his brain implant.

Graham and Philip O’Keefe, another ALS patient with paralysis, are the first two people to receive Synchron’s experimental brain-computer interface, which is inserted into a major vein in the neck rather than the brain. Doing so involves just a small incision into the neck — no brain surgery required, unlike other brain-computer interfaces. O’Keefe, 60, who got the implant in April, can now use a computer on his own to carry out work-related tasks and other activities.

Synchron’s CEO, Thomas Oxley, MD, PhD, describes the implant as “Bluetooth for the brain.”

“The big difference between what we have done with what previous approaches have done is that we do not put any needles into the brain or on top of the brain,” he tells Future Human. “The device stays inside the blood vessel.”

Felstead and O’Keefe just have to think about clicking a mouse or typing a letter and the computer does it for them. But learning how to use the system took extensive training with Synchron’s engineers. Three months after receiving the implant, they were able to use the device at their homes in Australia without any help. Now, Felstead and O’Keefe can type at about 14 and 20 characters per minute, respectively, without predictive text. Their click accuracy is between 92% and 93%. The two men are part of a small feasibility study, the results of which were published October 28 in the Journal of NeuroInterventional Surgery.

Unlike other implanted brain-computer interfaces that have been tested in people, Synchron’s device is delivered into the jugular vein. Surgeons use a catheter to thread the device upward through the vein until it’s adjacent to the motor cortex, the brain’s control center. The implant, a small mesh tube about an inch long, is dubbed the “Stentrode” for its resemblance to a heart stent, which is used to prop open clogged arteries.

The idea behind the device is to use the blood vessels as the natural highway into the brain. Like a heart stent, the Stentrode allows blood to flow freely through it.

Once implanted, the device picks up nearby brain activity using tiny electrodes that dot its surface. A second device, embedded in the patient’s chest, transmits the data wirelessly to a computer or other smart device running Synchron’s machine-learning software, which in turn converts that information into computer command functions.

Synchron thinks its device will be safe since heart stent surgery is a common procedure that’s been used for decades. Studies have shown they are generally safe and can last the rest of a person’s life. In a small number of people who get them, stents can be rejected and cause blot clots, also known as thrombosis. The other potential complication is that stents can migrate from their original location. This is a far more common side effect of stent surgery, occurring in about 20% of patients.

Oxley says Felstead and O’Keefe haven’t experienced either of these yet, but the company is monitoring them. While the device is designed to last a lifetime, Synchron has developed a procedure for removing the Stentrode if needed.

Other existing implanted brain-computer interfaces use chips with protruding metal spikes that are placed in or on the surface of the brain by removing a piece of a patient’s skull. The chips can cause brain inflammation and scarring, which eventually cause the device to stop working. As a result, current implants only last around five years.

Another major limitation of implanted brain-computer interfaces is that they can’t be used at home. The clunky setup involves a heavy pedestal that sits atop a person’s head and a thick cable to connect the implant to an external computer. These devices have allowed paralyzed people to type up to 10 words per minute or move robotic arms directly with their thoughts but only in a research lab.

Synchron thinks its device solves this problem and is hoping to be the first commercially available implantable brain-computer interface. Oxley says the company is now enrolling up to 15 patients in a clinical trial in Australia and plans to launch a study of six patients in the United States.

Robert Kirsch, PhD, a brain-computer interface expert and chair of biomedical engineering at Case Western Reserve University, who isn’t involved in Synchron’s work, says the company’s delivery approach is “elegant” for keyboard tasks, but the device isn’t exactly comparable to brain-computer interfaces designed to penetrate the brain or rest on its surface.

“This is a very creative and interesting approach for recording a relatively low number of brain signals with relatively low information content,” he tells Future Human. In other words, a simple task like moving a cursor or typing letters on a computer doesn’t demand much brainpower.

The brain-computer interfaces that require brain surgery typically record from 100 electrodes and can pick up signals from many individual neurons; Synchron’s device has just 16. (Elon Musk’s Neuralink says its implant will have 1,000 electrodes.) And since the Stentrode isn’t implanted in the brain, the number of these signals that it can pick up — known as the resolution — is likely far fewer than those detected by penetrating implants.

“Expanding the tasks much beyond the relatively simple keyboard tasks will be difficult in the current configuration,” Kirsch says.

Still, Oxley says the company wants to eventually use the Stentrode to help paralyzed patients control robotic limbs to give them even more freedom. That could mean fine-tuning the device and the software it uses to interpret brain signals into movements.

Even if the Stentrode is ultimately limited in what it can do, it could still have a positive impact on the everyday lives of the people who use it, like Felstead and O’Keefe.

“Using the Stentrode has been life-altering,” Felstead said. “It’s incredible to gain this level of independence back.”

Former staff writer at Medium, where I covered biotech, genetics, and Covid-19 for OneZero, Future Human, Elemental, and the Coronavirus Blog.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store