Science —

Building a bionic spine

Using a stentrode device, researchers want people to control robotic limbs via thought.

Behold, the stentrode.
Enlarge / Behold, the stentrode.
University of Melbourne

Australian neurologist Tom Oxley was on vacation in the US in November 2010 when he decided to do a bit of work. So he pitched an electrode array for reading brainwaves to the Defense Advanced Research Project Agency’s (DARPA) Military Prosthetics program. Instead of requiring invasive surgery to implant directly into the brain, Dr. Oxley’s electrode array could be threaded into a vein passing through the brain.

Oxley was surprised to get a response straight away, inviting him to discuss his device in person. At his meeting with DARPA’s Colonel Geoffrey Ling, Oxley got a bigger surprise still. Here’s how Oxley describes the encounter:

[Col. Ling] leaned back in his chair and said, “Well that’s interesting. No one has been doing that at DARPA. Why don’t you go home and put a team together, and we’ll give you a million dollars to build this thing.”

Easier said than done. As a neurologist, Oxley didn’t have the engineering know-how to build a complicated electronic device that could be compressed into a sub-millimeter catheter and still retain its signaling abilities. And he didn’t know who could; he wasn’t an academic with years of research collaborations and connections behind him. “It was surreal,” Oxley told one interviewer. “All of a sudden I’d started my PhD and had several million dollars to work with and a whole area to start my own lab.”

That turned out to be enough to start a large-scale collaboration that eventually produced a device called the stentrode. In February 2016, the stentrode was announced through a 39-author paper in Nature Biotechnology. How do you go from being a green graduate student to building both a research team and a working device in five years?

For Dr Oxley, the process started close to home. He was working at the University of Melbourne’s Parkville Precinct, which had developed both a bionic ear and a bionic eye. The university also had a school of medicine, as well as the Florey Institute of Neuroscience and Mental Health, for Oxley to draw on. Before he was done, Oxley’s team had spanned continents to combine disciplines as diverse as neurology, veterinary science, statistics, materials science, medical imaging, and electrical engineering. “I think it worked partly because I was young and open-minded,” Oxley says. “I never thought I had all the answers but that someone smarter than me did. So I kept on asking different academics till I heard what I thought was the most sensible solution.”

Bypassing brain surgery

Electrodes have been used to read brain activity and control external objects—it’s no longer the stuff of science fiction. Electrodes to the brain allowed a paralyzed man to kick off the 2014 World Cup and a quadriplegic to play Guitar Hero. But the technology is not without its drawbacks. For the best results, surgeons must open a patient’s skull and insert an electrode array in the brain itself, a procedure that comes with a 26-percent risk of infection or bleeding. The brain can also react to the electrodes with chronic inflammation and scarring, which makes the patient sick and reduces the electrodes’ ability to pick up signals. The dangers increase as surgeons try to access information-rich areas deep in the cortex of the brain.

While training to become a neurosurgeon, Oxley learned that stents can be inserted into blood vessels deep in the brain. He suspected that electrodes could be delivered in the same way. Stents are made from a biocompatible metal or plastic mesh that’s shaped into small pipes. These pipes are compressed and inserted into a catheter for delivery to a section of blood vessel that is blocked, either by a clot or by collapsed walls. Once ejected from the catheter, the stent expands and pushes open the walls of the vessel, allowing blood to flow freely again. Stents are often used to correct cardiovascular problems, such as cases where cholesterol plaques have reduced blood flow.

Oxley realized that a stent carrying electrodes would expand to press the electrodes against the walls of the vessel, putting them close to the brain tissue just outside. These electrodes could potentially read the activity of the nerve cells deep in the brain without requiring the invasive surgery currently in use.

Oxley’s stentrode is distinct from stents currently in use. At 3cm long, with crisscrossing metal struts and laser-cut platinum discs, the stentrode looks more like a delicate piece of jewelry than the frontier of medical technology. Human trials are set to begin in 2017, with the hope that paralyzed volunteers could use stentrodes to control their wheelchairs or even exoskeletons. “Essentially, what we’re trying to build is a bionic spine,” Oxley told Ars.

Like this, except much more bionic and modern.
Enlarge / Like this, except much more bionic and modern.
VintageMedStock/Getty Images

Step one: Paperwork

It wasn’t clear that any of that would be possible when Oxley first attracted DARPA’s interest. But his work began before he even left Washington DC, as he reached out to former classmate Rahul Sharma. Sharma is a cardiologist who also has business experience, so Oxley called him and said, “Rahul, DARPA are interested in funding an endovascular neural interface project. They suggested I go and get a patent immediately. We need cash." Oxley also turned to a friend who headed up an intellectual property firm in Melbourne who began working on a patent application. With the DARPA funding not yet official, Oxley and Sharma decided to cover the legal costs themselves.

Meanwhile, Oxley’s PhD supervisor, Professor of Medicine Terence O’Brien, kick-started the process of assembling a scientific team by introducing Oxley to engineers Tony Burkitt and David Grayden. Their work on the bionic ear and eye projects gave them in-depth knowledge of implanted devices, brain signaling, and signal processing. They in turn introduced Oxley to biomedical engineer Nick Opie, also from the bionic eye team. Opie eventually became Oxley’s co-leader for the project. Together, they spent the first half of 2011 working out a detailed proposal to present to DARPA in order to get their funding confirmed.

While he waited for the money, Oxley applied for permission to test the stentrode in animals in Australia. For this, he turned to the ethics committee of the Florey Institute of Neuroscience and Mental Health, where he was mentored by Clive May, head of the neurovascular group. May has spent a quarter-century studying how the brain controls the cardiovascular system, and he runs a hospital-style operating theater and intensive care unit for probing the brains of large animals.

“I got involved because Tom Oxley had nowhere to do the studies, no experience in the surgical techniques required, and no experience in using large animals,” May told Ars. Together, he and Oxley planned how the still-theoretical stentrode could be tested for safety and efficacy in a living creature.

Channel Ars Technica