Skip to main content

31-07-2012 | Neurology | Article

Low-cost device affords motor deficiency patients gaze-control


Free abstract

MedWire News: Researchers have developed a low-cost, 3D gaze estimation device to complement direct brain-machine interfaces (BMI), which could allow patients with serious motor deficiencies to interact with computers and their surroundings.

This new device is composed from inexpensive off-the-shelf materials and enables a computer to work out exactly where a person is looking by tracking their eye movements, thus allowing them to control a cursor on a screen like a conventional computer mouse.

Indeed, the GT3D system has solved the "Midas touch problem" of differentiation between normal behavioral eye movements and intentional eye "commands" by using a simple wink to represent a mouse click.

Commenting on the study findings in a press release, lead study author Aldo Faisal (Imperial College London, UK) said: "Crucially, we have achieved two things: we have built a 3D eye tracking system hundreds of times cheaper than commercial systems and used it to build a real-time brain machine interface that allows patients to interact more smoothly and more quickly than existing invasive technologies that are tens of thousands of times more expensive."

Using mass-produced video-game hardware, the researchers created a device made up of two fast video game console cameras attached to a pair of glasses, with a total cost of US$ 30 (€ 24.50). The device is 800 times cheaper than conventional BMI systems and operates at a sampling rate of over 120 Hz with a 0.5-1 degree of visual angle resolution.

The cameras attached to the glasses constantly take pictures of the eyes to work out where the pupils are pointing, and from this the researchers were able to use a set of calibrations to work out exactly where a person is looking on a screen.

Furthermore, the team was able to use more detailed calibrations to identify the 3D gaze of participants. This could allow people to control an electronic wheelchair or a robotic prosthetic arm by looking where they want to go.

The effectiveness of the GT3D device was assessed by asking six participants to play the "pong" computer game using only the eyes to move a line to hit a ball bouncing around the screen. On average, individuals achieved a score of 6.6 with the GT3D compared with 8.3 using the mouse, following first-time use of the device for only 10 minutes.

"Our approach, unlike other BMI technologies, enables us to use gaze information to infer user intention in the context of its natural occurrence… This approach drastically reduces training time and boosts patient adherence," conclude the researchers in the Journal of Neural Engineering.

MedWire ( is an independent clinical news service provided by Springer Healthcare Limited. © Springer Healthcare Ltd; 2012

By Ingrid Grasmo

Related topics