On Diagnosing Autism, the Eyes Have It

Infrared eye-tracking technology could aid in early autism diagnoses, even in non-verbal children, which can mean more effective treatment.


Infrared eye-tracking technology could unlock rapid and straightforward autism diagnosis that works on young children, whether they are verbal or not.

Compared to neurotypical children, autistic children tend to look at faces in a different way. For instance, they tend to look at mouths more and at eyes less.

Those are the findings of a new University of Waterloo-led study that specifies an algorithm that is ready to be implemented by software developers. Mathematicians Anita Layton and Mehrshad Sadria published their study in Computers in Biology and Medicine.

Early diagnosis of autism can help kids get behavioural interventions and speech therapy sooner, and that makes these treatments more effective. But conventional testing involves complex questionnaires or lengthy evaluations by a psychologist, which can be expensive and stressful for children. They may also struggle to focus for long enough to get an accurate result.

The authors hope that their approach will provide an objective and less error-prone diagnosis in just a few minutes where children simply look at images of faces on a screen. Data collected from tracking their gaze using a commercially available infrared device could be used to calculate a diagnosis.

“We use mathematics as a microscope to understand biology and medicine,” said Layton and Sadria in their article in The Conversation.

The study compared the gaze patterns of autistic and neurotypical children around the age of 5 using a set of 44 photographs of faces on a 19-inch screen. An infrared device was used to track where the child was looking based on emission and reflection signals from the iris.

Unlike previous eye-tracking studies that looked only at time spent fixated on various facial features, the authors also looked at the paths the eyes take between features. The study specified a set of seven areas of interest: under the right eye, right eye, under the left eye, left eye, nose, mouth and other parts of the screen.

For each participant, the time spent fixated on each area was calculated, and how the eyes moved between them was analyzed using concepts from network analysis. The autistic children spent more time looking at mouths, and they scanned between facial features more slowly, their eyes taking a different characteristic path than their neurotypical peers.

This novel combination of fixation time and transition speed and patterns gives a more robust evaluation of gaze, and it can be done in minutes.

Being able to diagnose autism earlier allows therapy to start sooner, a key factor in how effective treatment can be. Further development and wider validation of this method could help people with autism lead more independent lives.

‹ Previous post
Next post ›

Karyn Ho is a science animator and engineer who thrives at the interface between science, engineering, medicine, and art. She earned her MScBMC (biomedical communications) and PhD (chemical engineering and biomedical engineering) at the University of Toronto. Karyn is passionate about using cutting edge discoveries to create dynamic stories as a way of supporting innovation, collaboration, education, and informed decision making. By translating knowledge into narratives, her vision is to captivate people, spark their curiosity, and motivate them to share what they learned.