A brain cap paired with intelligent algorithms could potentially allow paralyzed patients to translate their thoughts into movement—without the need for surgery.
Individuals with spinal cord injuries frequently lose partial or full movement in their arms or legs. Often, the nerves in the limbs remain functional, and the brain still generates normal signals. The issue lies in the spinal cord injury, which disrupts communication between the brain and body.
Scientists are now investigating methods to restore these signals without directly repairing the spinal cord.
Decoding Movement Intent Through Brain Scans
In a study published today (January 20) in APL Bioengineering by AIP Publishing, researchers from universities in Italy and Switzerland investigated whether electroencephalography (EEG) could connect brain activity to limb movement. They explored whether this noninvasive technology could detect the brain’s movement signals and make them functional again.
When a person tries to move a paralyzed limb, the brain still generates the same electrical patterns as it would for normal movement. If these signals can be captured and interpreted, they could be relayed to a spinal cord stimulator, which might then activate the nerves controlling that limb.
Reasons to Avoid Brain Implants
Much of the previous work in this area has used surgically implanted electrodes to directly capture movement-related brain signals. While effective, the researchers aimed to determine if EEG could provide a safer option.
EEG systems consist of caps with multiple electrodes that record brain activity from the scalp. Though they may seem complex, the team notes they carry far less risk than inserting hardware into the brain or spinal cord.
“Implants can lead to infections and require additional surgery,” explained author Laura Toni. “We wanted to see if that risk could be avoided.”
Challenges of EEG Technology
Detecting movement signals with EEG comes with major technical challenges. Since the electrodes rest on the scalp, they struggle to pick up activity from deeper brain regions, making some movements harder to read than others.
Signals for arm and hand movements are easier to capture because they originate near the brain’s surface. In contrast, signals for leg and foot movements are more difficult to decode since they arise from deeper, more central areas.
“The brain controls lower limb movements mainly in the central region, while upper limb movements are closer to the surface,” explained Toni. “This makes it easier to map and decode signals from the upper limbs than from the lower ones.”
Using Machine Learning to Interpret Brain Signals
To interpret the limited EEG data, the researchers applied a machine learning algorithm capable of handling small, complex datasets. During experiments, participants wore EEG caps while trying to perform simple movements. The team captured the resulting brain activity and trained the algorithm to organize and classify the signals.
The system could consistently detect when a person intended to move versus when they did not, but it had difficulty differentiating between specific types of movements.
Next Steps
The researchers believe their method can be enhanced through further development. Future efforts will aim to improve the algorithm’s ability to recognize specific actions like standing, walking, or climbing, and to investigate how these decoded signals might trigger implanted stimulators in patients recovering from injury.
If successful, this approach could bring noninvasive brain scanning closer to enabling people with spinal cord injuries to regain functional movement.
Read the original article on: SciTechDaily
Read more: AI Glasses Help the Visually Impaired Navigate Safely