New Technology Developed to Detect Autism in Children

16 July 2019


A new method to diagnose autism spectrum disorder (ASD) in children has been developed by a team of researchers at the University of Waterloo in Ontario, Canada.

Current methods of diagnosis involve surveys and psychological evaluations that can sometimes be difficult to administer to young children. The new technology however analyses the child's visual gaze and offers an easy, less stressful test.

University of Waterloo Applied Mathematics, Pharmacy and Biology professor Anita Layton said: “It is much easier for children to just look at something, like the animated face of a dog, than to fill out a questionnaire or be evaluated by a psychologist.

“Our technique is not just about behaviour or whether a child is focusing on the mouth or eyes. It’s about how a child looks at everything.”

A recent study carried out by the team has shown that children with ASD scans a person's face differently to that of a “neuro-typical” child, and move their eyes around facial features in a very distinctive way.

During the study, 40 four to five year olds, of which 17 were ASD children, were shown 44 photographs on a screen. An infrared eye-tracking device was used to analyse how their eye moved between seven significant areas of interest (AOIs) - under the right eye, the right eye itself, under the left eye, the left eye itself, the nose, mouth and then finally other parts of the image. It was found that the gaze of children with ASD moved between the AOIs quicker and spent more time viewing the mouth, while spending less time on the eyes.

University of Waterloo Applied Mathematics department master’s student Mehrshad Sadria said: “Many people are suffering from autism, and we need early diagnosis, especially in children.

“The current approaches to determining if someone has autism are not really child-friendly. Our method allows for the diagnosis to be made more easily and with less possibility of mistakes.”

The team hope to refine the process to combine it with traditional methods of diagnosis.