Brain-Computer Interfaces Evolve to Help People With Paralysis
All your movements start out in your brain.
When you decided that you wanted to read this article, you planned on moving your finger (or your cursor) toward a certain spot on your screen. Without noticing it, you thought about pressing or clicking on that spot. After quickly processing that thought, your brain told your muscles to respond to it accordingly, and here you are.
If the damage is permanent, so is paralysis. In these cases, adaptation is essential to improve patients’ lives. This is why, many scientists are working on new assistive technologies.
Brain-computer interfaces (BCI) are one of the most promising innovations in this field. Through electrodes, BCIs can read and translate brain signals into commands for an output device that can carry out the user’s intention. This way, people with restricted motor functions can move robotic prostheses or use computers with their minds.
How do brain-computer interfaces work?
The brain uses specialized cells called neurons to carry messages. Whenever we think, small electrical signals are carried from neuron to neuron. These signals are generated by differences in electric potential carried by ions on the membrane of each neuron. These signals can be detected using electrodes or a device called an electroencephalograph (EEG). These devices measure the tiny differences in the voltage between neurons. These differences are then interpreted by a computer algorithm and can be used to direct computers or prostheses.
There are several methods used to collect electrical signals from the brain and transmit them to computers. These methods include:
- Non-invasive. BCI’s electrodes measure brain activity through the scalp. There is no need for surgery and the device is visible to the naked eye.
- Semi-invasive. BCI’s electrodes are installed via craniotomy on the exposed surface of the brain, such as the dura mater or the arachnoid mater.
- Invasive. BCI’s electrodes are surgically implanted into the cortex of the brain. These are the most effective devices because they produce the highest quality signals, but they increase the risk of scar-tissue build-ups around the electrodes.
Specific techniques for establishing communication between brain and machine are being researched by different neural engineering companies.
Wireless BCI systems
Everything is going wireless nowadays. We have wireless Internet connections, wireless headphones, wireless keyboards. So why wouldn’t we also have practical, wireless BCI systems?
Neuralink, one of the most famous neural engineering firms, aims at building a BCI that links brains and computers via Bluetooth. In August 2020, Neuralink CEO Elon Musk conducted a public demonstration of Link VO.9, an implantable, coin-sized chip with 1024 electrodes that registers neural activity aided by microscopic threads.
The chip was inserted into a pig’s brain with highly precise robotic surgery. In the demonstration, it was able to predict movement and measure temperature and intracranial pressure in real time — something that Musk stated could help predict strokes or heart attacks.
According to its website, Neuralink’s main mission is to help people with spinal cord injuries and neurological disorders by recording the activity of thousands of neurons in the brain. The Link is meant to receive and decode that information, and then send it to the users’ computer to allow them to control virtual mouses, keyboards, and even game controllers. How well this works, will all depend on the improvement of the decoding algorithms.
BrainGate, another neurotechnology company in the U.S., recently tested a high-bandwidth intracortical BCI on humans that delivers brain signals using external wireless transmitters instead of cables.
The transmitters were placed on top of the user’s head and connected to sensors inside the brain through the same port used by wired devices. Employing the same decoding algorithms as wired BCIs, the wireless device performed as well in the clinical trials as the wired BCIs, providing high-fidelity signals and similar accuracy in the patients’ control over the computer.
In clinical trials, two people with spinal cord injuries were connected to a standard tablet computer via BrainGate’s BCI. Researchers evaluated their point-and-click precision and typing speeds. Because they weren’t limited by cables, the patients found it easier to use BrainGate’s BCI for longer periods.
“We want to understand how neural signals evolve over time,” said Leigh Hochberg, an engineering professor at Brown University who led the BrainGate clinical trial. “With this system, we’re able to look at brain activity, at home, over long periods in a way that was nearly impossible before. This will help us to design decoding algorithms that provide for the seamless, intuitive, reliable restoration of communication and mobility for people with paralysis.”
Stent-electrode recording array
Also known as the Stentrode, this device was developed by a team of the University of Melbourne. It consists of an electrode array mounted on a tiny stent. Just like a normal intracranial stent, the device is implanted into a blood vessel in the brain with a catheter.
The main advantage of this method is that open brain surgery is replaced by a tiny incision in the neck.
In 2020, the Stentrode was successfully tested on two patients with motor neuron diseases, who managed to control a computer-based operating system via an eye-tracker for cursor navigation. This way, the trial participants managed to use text, email, and conduct online shopping only with their minds.
Neurograins
Neurograins — as dubbed by their creators at Brown University — are silicon-based neural sensors about the size of a grain of salt. These microscale chips record the electrical pulses of neurons and, because they’re spread across the brain, they’re able to transmit a massive amount of data to an external central hub.
The main advantage of this BCI system is that it covers many different points in the brain. “Up to now, most BCIs have been monolithic devices — a bit like little beds of needles. Our team’s idea was to break up that monolith into tiny sensors that could be distributed across the cerebral cortex”, explained Arto Nurmikko, a professor in Brown’s School of Engineering who leads the investigation.
The central hub is a patch placed on the scalp that uses a network protocol to coordinate the signals individually (each neurograin has its own network address). It also sends a tiny amount of electricity to the neurograins to power them wirelessly from outside the skull. Similar electrical pulses can be sent to stimulate neural activity. Researchers hope this can help people with paralysis recover brain function, and also treat people with Parkinson’s disease and epilepsy.
So far, however, the neurograins have only been tested on rodents, whose small brains only require 48 grains — while human brains would need around 770.
Other applications of brain-computer interfaces
Although the main BCI studies have medical motivations, German automobile manufacturer Mercedes-Benz recently published a press release about it is incorporating BCI technologies that its Vision AVTR concept car will include. The carmakers' ultimate goal is to simplify vehicle operation and biometric interaction by controlling the user interface through brain activity.
In other words, you could wear a headset when driving and switch on the radio and the lights — and someday even drive the car — with your mind.
On the other hand, Elon Musk has declared that Neuralink’s devices could also be used by healthy people in the future. If the BCI devices learn to communicate with other areas in the brain, they could have other applications besides medicine. For instance, BCIs could be used for Augmented Intelligence, a subsection of machine learning focused on improving human cognition with the assistance of AI.
Post a Comment