Researchers develop new brain-computer interface that lets paralyzed patients use tablets

Researchers have developed a new iBCI (intracortical brain-computer interface) that allows people with paralysis to control an unmodified, commercially available tablet. This research was based on the fact that most general-purpose computers have been difficult to use for people with some form of paralysis.

In their study, three research participants with tetraplegia who had multielectrode arrays implanted in motor cortex as part of the BrainGate2 clinical trial were invited. Using the iBCI, their neural activity was decoded in real time with a point-and-click wireless Bluetooth mouse. This allowed participants to use common and recreational applications (web browsing, email, chatting, playing music on a piano application, sending text messages, etc.). iBCI also allowed two participants to “chat” with each other in real time.

The architecture of the setup

Participants used seven common applications on the tablet: an email client, a chat program, a web browser, a weather program, a news aggregator, a video sharing program, and a streaming music program.

The system consisted of a NeuroPort recording system to record neural signals from the participant’s motor cortex.

These signals were routed into a real-time computer running the xPC/Simulink Real-Time operating system for processing and decoding.

The output of the decoding algorithm was passed to a Bluetooth interface configured to work as a wireless computer mouse using the Bluetooth Human Interface Device (HID) Profile.

This virtual Bluetooth mouse was paired with a commercial Android tablet device with no modifications to the operating system.

Participants performed real-time “point-and-click” control over a cursor that appeared on the tablet computer once paired through the Bluetooth interface.

The cursor movements and clicks by participants were decoded from neural activity using Kalman filters. 2D cursor velocities were estimated using a Recalibrated Feedback Intention Trained Kalman Filter (ReFIT-KF) and a cumulative closed-loop decoder. Click intentions were classified using a hidden Markov model and a linear discriminant analysis classifier.