EEG technology could be a promising tool for connecting brain signals with limb movements, offering a potential solution for individuals with spinal cord injuries. These injuries often result in the loss of limb function, despite the nerves in the limbs and neurons in the brain remaining intact. The challenge lies in the inability of the damaged spinal cord to facilitate communication between these two areas.
A recent study published in APL Bioengineering by AIP Publishing delves into the feasibility of using electroencephalography (EEG) to bridge this gap. When a patient attempts to move a paralyzed limb, the brain generates specific signals corresponding to that movement. By decoding these signals, they can be transmitted to a spinal cord stimulator, which then controls the nerve endings in the affected limb.
The research team, comprising scientists from Italian and Swiss universities, chose to explore EEG technology due to its non-invasive nature. Unlike implantable electrodes, which have shown some success, EEG devices are typically worn as caps adorned with numerous electrodes that measure brain activity. This approach avoids the risks associated with surgical procedures, such as infections, as noted by author Laura Toni.
However, decoding limb movements using EEGs presents technical challenges. The surface-placed electrodes may struggle to capture signals from the brain's deeper regions, which is more problematic for leg and foot movements compared to arm and hand movements. Toni explains that the brain's control over lower limb movements is primarily centralized, making it more complex to map and decode.
To address this, the researchers employed a machine learning algorithm designed for limited datasets. In their tests, patients wore EEG monitors and performed simple movements, allowing the algorithm to classify the signals. While they successfully differentiated between attempted movement and no movement, distinguishing between specific signals proved more difficult.
Looking ahead, the team aims to enhance their algorithm's ability to recognize various movement attempts, such as standing, walking, or climbing. They also plan to explore ways to utilize this data to trigger these movements in spinal cord injury patients, potentially improving their mobility and quality of life.
