In the context of biomedical engineering, the goal of BCIs is to provide functional support systems for people with disabilities, to replace or recover lost functions. Using a noninvasive brain-computer interface, researchers of Carnegie Mellon University, in collaboration with the University of Minnesota, have developed an innovative system to control a robotic arm with the mind. The technology can be applied to patients without any surgery, avoiding the associated risks.
In the near future, the system could benefit patients with paralysis and those with movement disorders.
An important step in noninvasive BCI
Previously, brain-computer interfaces that use signals acquired with intracortical implants have achieved successful results for neural control of robotic devices. But, since intracortical implants require invasive surgical operations that could involve considerable risks, finding a noninvasive alternative that can provide high-quality control would improve the integration of BCIs into clinical settings. However, there’s a problem which needs to be addressed. BCIs that use noninvasive external sensing, compared to brain prostheses, receive signals disturbed by noise, which leads to a lower resolution and less precise control. The system developed by the researchers of Carnegie Mellon University has proven to be able to overcome this problem.
The novel approach is based on EEG and allows to achieve the neural control of a robotic arm for continuous random target tracking. This result has been reached by developing novel machine learning techniques which allowed access to signals deep in the brain, obtaining a high resolution of control on a robotic device.
A novel approach
To demonstrate the effectiveness of their technology, the researchers tested the device on 68 able-bodied subjects. During the tests, the robotic arm was able to continuously track a cursor in real-time.
In the paper, “Noninvasive neuroimaging enhances continuous neural tracking for robotic device control”, published in Science Robotics, researchers introduced a new framework that addresses and enhances upon both the “brain” and “computer” components by increasing not only user engagement through a continuous pursuit task and associated training paradigm, but also the spatial resolution of noninvasive neural data through EEG source imaging.
The test results are remarkable: the study demonstrates that the new approach enhanced BCI learning by nearly 60% for traditional center-out tasks and by more than 500% in the more realistic continuous pursuit task. According to the team, the device is ready to be tested in clinical trials.
“This work represents an important step in noninvasive brain-computer interfaces, a technology which someday may become a pervasive assistive technology aiding everyone, like smartphones.” – Bin He, Department Head, Biomedical Engineering.
The work was supported in part by the National Center for Complementary and Integrative Health, National Institute of Neurological Disorders and Stroke, National Institute of Biomedical Imaging and Bioengineering, and National Institute of Mental Health.
If you want to know more about mind-control of robotic arm read also: Multitasking: a BMI to control a third robotic arm