New detection system developed that analyzes gaze and detects autism

A new method to more effectively detect autism spectrum disorders in children has been developed by researchers at the University of Waterloo. The new method is based on the analytical and detailed observation of the gaze of children in front of people’s faces.

The researchers, who published their own study on Computers in Biology and Medicine, noticed that children with an autism spectrum disorder scan the face of people differently than children without autism.

In the course of the experiments, the researchers analyzed the reactions of 17 children with an autism spectrum disorder and 23 neurotypical children when they were shown photographs of faces on a screen. The eyes of the children were traced by a special infrared system that also analyzed the iris.

By analyzing how the children moved their eyes and scrutinized the faces as well as the ways in which the eye moved, the researchers were able to establish the presence of an autism spectrum disorder. It is a method that according to the authors could be very useful also because the current approaches, as specified by Mehrshad Sadria, one of the authors of the study, are not very suitable for children and are subject to errors.

“Our technique is not just about behavior or whether a child concentrates on the mouth or the eyes, but it is about how a child looks at everything,” says Anita Layton, professor of applied mathematics, pharmacy and biology at Waterloo and another author of the study.