A brain cap and algorithms may one day let paralysed people trigger limb movement without surgery, researchers in Italy and Switzerland say.
People with spinal cord injuries can lose some or all movement in their arms or legs. In many cases, the nerves in the limbs still work and movement signals are still produced.
The problem is that damage to the spinal cord blocks communication between the brain and the rest of the body, so researchers are exploring ways to reconnect those signals without repairing the spinal cord itself.
Scientists at universities in Italy and Switzerland examined whether electroencephalography, or EEG, could help link brain activity to limb movement.
EEG systems are worn as caps fitted with multiple electrodes that record activity from the scalp.
When someone attempts to move a paralysed limb, their signals are still produced.
If these signals can be detected and interpreted, they could be sent to a spinal cord stimulator, a device that sends electrical pulses to nerves, which may then activate the nerves responsible for movement in that limb.
Much of the earlier research in this area has relied on surgically implanted electrodes to read movement-related signals. Although those systems have shown promise, the researchers wanted to see if EEG could offer a safer alternative.
Author Laura Toni said: “It can cause infections; it’s another surgical procedure. We were wondering whether that could be avoided.”
Reading movement signals through EEG presents technical challenges.
Because the electrodes sit on the surface of the head, they can struggle to detect activity that starts deeper inside.
Signals related to arm and hand motion are easier to detect because they originate closer to the outer regions.
Movements involving the legs and feet are harder to decode because those signals come from areas located deeper and closer to the centre.
Toni said: “The brain controls lower limb movements mainly in the central area, while upper limb movements are more on the outside.
“It’s easier to have a spatial mapping of what you’re trying to decode compared to the lower limbs.”
To make sense of limited EEG data, the researchers used a machine learning algorithm designed to analyse small, complex datasets.
During testing, patients wore EEG caps while attempting simple movements. The team recorded the activity produced during these efforts and trained the algorithm to sort and classify the signals.
The system was able to reliably tell when a person was trying to move versus when they were not. However, it struggled to distinguish between different types of movement attempts.
The researchers believe their approach can be improved with further development.
Future work will focus on refining the algorithm so it can identify specific actions such as standing, walking or climbing.
They also hope to explore how these decoded signals could be used to activate implanted stimulators in patients undergoing recovery.
