Good Thinking

MagTrack system uses face and head movements to control devices

View 2 Images
The prototype MagTrack system, pictured here minus its hard-wired tongue sensor
Georgia Tech
The prototype MagTrack system, pictured here minus its hard-wired tongue sensor
Georgia Tech
The MagTrack IMUs are temporarily attached via a non-toxic biocompatible adhesive
Georgia Tech

Although there are various systems that detect quadriplegics' head movements, most such setups are limited to the control of wheelchairs. Such is not the case, however, with the MagTrack system.

The technology is being developed via a partnership between the Georgia Institute of Technology and US company Brooks Rehabilitation.

In its present form, MagTrack incorporates a pair of "glasses" – really just frames with no lenses – along with multiple IMUs (inertial measurement units). Each IMU detects body movements via a combination of an accelerometer, gyroscope and magnetometer.

The glasses themselves contain some IMUs, while additional hard-wired facial units are temporarily adhered to the user's right-hand eyebrow and cheek, and even their tongue. What's more, the glasses incorporate a row of magnets, the magnetic field of which is measured by the IMUs – this allows each facial IMU to "know" where it currently is, relative to the glasses.

The MagTrack IMUs are temporarily attached via a non-toxic biocompatible adhesive
Georgia Tech

As the user moves their eyebrow, cheek or tongue, the relevant IMU detects the movement. Additionally, if the person tilts their whole head forward, backward or to either side, the IMUs within the glasses detect that movement, too.

In all cases, the detected action is used to trigger a function of a paired electronic device, via a wireless programmable control unit. Should that device be a powered wheelchair, then the user's movements can be used to start, stop and steer it.

Given the fact that the system is portable, however, the person can also use it when not in their wheelchair, to control devices such as smartphones, computers or smart TVs. For those items, the user's movements can be utilized to perform actions like scrolling, mousing and selecting.

The technology has already been tested by 17 wheelchair users, who became proficient in its use after less than three hours of training. In fact, the system allowed them to perform wheelchair-driving tasks either as fast or faster than was possible via their conventional controllers.

A paper on the system was recently published in the journal IEEE Transactions on Biomedical Engineering. The scientists are now working on a less conspicuous, more capable version of MagTrack.

"The trajectory of the MagTrack study shows an unprecedented possibility for the advancement of independent function as well as mobility for electric wheelchair users," says Georgia Tech's Prof. Omer T. Inan, who is leading the research along with co-inventors Nordine Sebkhi and Arpan Bhavsar. "Our team and partners are energized and motivated by the recent patient trials to continue to push this technology and its capabilities as far as possible."

Sources: Georgia Tech, MagTrack

  • Facebook
  • Twitter
  • Flipboard
  • LinkedIn
1 comment
freddotu
It's encouraging to see something that appears affordable (3d printed frames, off the shelf IMUs, etc) and is an improvement on current assistive technologies. Considering the number of Arduino, Raspberry Pi and 3d printing enthusiasts out here, one could hope this to be open source and shared with as many as need it. I'd be more than happy to build a handful of sets to distribute.