Under the direction of Woon-Hong Yeo at the Georgia Institute of Technology, a multi-institutional research team has developed a wireless wearable brain-machine interface (BMI) system using soft scalp electronics. The new system could make BMI technology more comfortable and easier to use while improving the accuracy of brain signal capture.
A brain-machine interface offers enhanced rehabilitation for neuromuscular conditions that cause motor dysfunction. Currently, most BMI systems use wired electroencephalography (EEG) brain mapping technology that requires hard electrodes, gels, pastes, and a tight skullcap to secure electrodes to the scalp.
In a departure from these cumbersome BMI systems, the international team built a portable EEG system that combines virtually imperceptible microneedle electrodes with soft wireless electrical circuits. The EEG wearable could allow a user to operate a wireless robotic arm or wheelchair by imagining the action in their mind.
Standard BMI systems require a tangle of multiple wires; they are time consuming to set up and require gels or pastes to affix the hard electrode sensor to the skin. Maintaining proper electrode contact is an ongoing challenge, and displaced electrodes degrade signal acquisition, leading to reduced accuracy.
In contrast, the new BMI is easier and faster to apply, requires no gels, and offers a more comfortable experience because the electrodes and circuitry are soft and flexible. There’s also less chance of signal noise because the electrodes don’t shift around so easily, leading to improved accuracy.
The researchers further improved accuracy by using VR technology and powerful AI algorithms to train the BMI. In preliminary tests, four non-disabled human subjects successfully controlled movements in a VR environment using imagined motor actions, such as grasping.
Although still in the early stages, the new BMI holds promise for individuals with severe paralysis and locked-in syndrome, a condition in which a fully conscious individual cannot move or communicate. Further development and testing is planned with subjects across a range of motor dysfunction conditions. The researchers published their findings in the journal Advanced Science.