How engineers are trying to build better brain implants

Although Elon Musk’s startup Neuralink might have you thinking that brain-computer interfaces, or BCIs, are dystopian, futuristic devices, the technology has actually been around for decades.

BCI technology links your brain and a computer, acting as a window into the mind. It works by collecting signals from the brain, analysing them and then transforming them into commands; BCIs have allowed quadriplegic patients to control computers, neuroprostheses and exoskeletons. The technology has increasingly grown more sophisticated in recent years, and an injection of cash into the commercialisation of the technology promises leaps and bounds ahead. 

But there are myriad hurdles to overcome. There are several important unanswered questions that BCI research must address, and one of the biggest ones is how long can an implant safely and effectively last in a patient. “One of the big goals is making them last as long as possible,” says Chad Bouton, director of the Neural Bypass and Brain-Computer Interface Laboratory in The Feinstein Institutes for Medical Research at Northwell Health in New York.

The brain is surprisingly amenable to change – including the insertion of a chunk of metal. But the challenge is keeping it in without damaging brain tissue, causing a chronic immune response or the implant itself degrading. All of which not only poses a safety risk to the patient, but also affects the quality of the signal the device can transmit. 

A man called Nathan Copeland, who was paralyzed from the chest down and fitted with a BCI – he is perhaps best known for fistbumping Barack Obama with his robotic hand – currently holds the record for the longest an implant has stayed in a patient: for Copeland, that’s about 9 years so far. The previous record holder was a man named Ian Burkhart, whose research was led by Bouton. Burkhart kept his BCI in for six years, but had it taken out in August 2021 when an infection developed where the cable led into his scalp. 

The most commonly-used BCI technology is called the Utah array, made by Blackrock Neurotech. “It's one of the most mature technologies; it’s been around for a very long time,” says Bouton. It looks like a tiny square, smaller than a coin, with 100 tiny spikes protruding from it, between 0.5 and 1.5mm long. Dozens of patients around the world have had one implanted. So far, the Utah array has had commercial FDA clearance, but only up to 30 days or less, although Blackrock is pursuing clearance for longer. The arrays are made of biocompatible materials, meaning materials that don’t mess with human tissue. The needles themselves are made of silicon, which is metalized – most commonly platinum or iridium oxide – and then insulated with a material called parylene C. 

While it’s the most popular, implanting the Utah array is still a pretty invasive procedure. Other companies and research groups are taking a different tack. Musk’s Neuralink relies on thin, flexible threads – thinner than human hair – that record neural activity across over 1,000 electrodes, although implantation requires drilling a hole in the skull. (In May, it was revealed that Neuralink’s first human volunteer had experienced issues with his device that may be due to its unique design). Neuralink’s competitor Synchron, a US-based startup, avoids brain surgery altogether, getting its technology into the brain via a stent, a metal scaffold, and the device is placed in a blood vessel in the motor cortex.

Taking an even more minimally invasive approach is Motif Neurotech, founded by Jacob Robinson, a professor of engineering at Rice University. Robinson and his team at Motif are developing a pea-sized device that penetrates and collects signals from the skull, meaning it doesn’t need to come into contact with brain tissue and is safer. The tradeoff is that it collects fewer signals, but Robinson says they don’t need as many to do what they want to do, which is treat mental health conditions. “We are approaching the problem by thinking about what is the best way to put a device in the body so that it can get the information that is really critical for us to provide therapeutic benefit – but not produce significant biological response,” says Robinson. 

It’s not risk-free – it can still cause a scalp infection at the point of attachment – but Robinson reckons it’s significantly safer than other approaches. The implant has special material they specially designed, called magnetoelectric material, which vibrates in the presence of the magnetic field which then turns into data and power. The rest of the implant is made of materials used in other biomedical devices. So far, they’ve implanted the device in two people and are moving to early feasibility trials to implant the device for longer to see how it fares.  

Robinson says there are two ways of getting around the problem of how the brain will naturally react to foreign material: coming up with different ways to collect the signals from the brain, or changing the materials that make up the device. Most of the experimentation in the field appears to be focusing on the latter. 

And for now, no one can answer exactly how long a BCI can last. “The longevity of BCI remains somewhat unknown,” says Robinson. Only more research – and time – will tell. 

This project was funded by:

Through the Health Innovation call.