Machine Design

A Tongue-Activated Sensor Lets Patients Paralyzed From the Neck Down Maneuver a Wheelchair or Run a Computer Program

Georgia Institute of Technology,

Patients with high-level spinal injuries — those who cannot move their arms, legs, and sometimes even their heads — can have a hard time controlling wheelchairs or running computer programs, even with the adaptive mechanisms developed by biomedical engineers. A new device, the Tongue Drive, could let such individuals control a variety of machines, including wheelchairs and computers.

The drive consists of a grain-of-rice-sized magnet attached to a patient’s tongue with tissue adhesive plus a wireless headphone containing an array of magnetic field detectors. The array detects movements of the magnet on the tongue and sends those tracking signals to a computer stowed on the wheelchair. The computer processes the signals to determine the relative motion of the magnet. This information controls the cursor on a computer screen or substitutes for a joystick on a motorized wheelchair. Tests show that users can quickly train the computer to recognize tongue movements.

The inventor, Maysam Ghovanloo, chose the tongue as a means to operate the controller because it connects directly to the brain. And the tongue is rarely affected by spinal-cord injuries. Although the initial trial involved just six different commands, the Tongue Drive can be trained to identify hundreds of tongue movements, probably more than a patient can execute. And each movement can represent a different command. This scheme is much more flexible than traditional sip-n-puff devices that let patients communicate by either exhaling or inhaling to control a simple switch. The inventor is currently working on a design that places the magnet array inside the patient’s mouth, which should help those so impaired they cannot position the headset for the best results.

TAGS: Sensors
Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.