iPad Play Could Help Identify Signs of Autism

September 7, 2016

A clinical diagnosis of autism spectrum disorder (ASD) is time consuming, and requires detailed evaluation and physician expertise. No single lab test can detect ASD, since symptoms are extremely broad and unique to each individual. Typically, a developmental pediatrician, psychiatrist, or neurologist will use a diagnostic tool like the modified checklist of autism in toddlers (M-CHAT) to determine whether a child needs further evaluation. Then, the professional will complete a series of behavior observations focused on the social and emotional aspects of ASD.

Now, research published in Scientific Reports suggests there might be an easier way to detect early indicators of ASD. Scientists from the University of Strathclyde and Harimata used tablets with touch-sensitive screens and embedded inertial movement sensors to evaluate 37 children with autism between 3 and 6 years old. For the study, the researchers adapted two commercially available games – Sharing and Creativity – to capture inertial sensor data and touch-screen data as the kids played. They compared the collected kinematics and gesture forces to 45 age- and gender-matched children without autism to determine if movement patterns reliably differentiated the children, then identify which kinds of movements were different.

Three machine learning algorithms that focused on motor patterns identified the children with autism with 93% accuracy. The children with autism had a particular motor signature while playing that was different from the pattern produced by children without ASD. This suggests that motor disturbance is a new target for early assessment and diagnostic instruments.

"This is potentially a major breakthrough for early identification of autism, because no stressful and expensive tests by clinicians are needed,” says Dr. Delafield-Butt, a senior lecturer in child development at Strathclyde’s School of Education. “Early detection is important as this can allow parents and children to gain access to a range of services support.”

The research team hopes that its findings will reduce the clinically demanding burden on physicians, and allow for earlier symptom identification, since goal-directed motor control deficits are observable from shortly after birth. Previously, motion tracking required expensive, lab-based systems, but inertial motion sensors, gyroscopes, and magnetometers that are common in consumer tablets give access to motor information that could be used for medical assessment.

This is the first study to use tablet games to study movement patterns of children with ASD, and researchers suspect the motor signature of ASD may overlap with other developmental coordination disorder or ADHD. More research is needed to fully understand the patterns.

Journal Reference:

1. Anna Anzulewicz, Krzysztof Sobota & Jonathan T. Delafield-Butt. Toward the Autism Motor Signature: Gesture patterns during smart tablet gameplay identify children with autism. Scientific Reports, August 2016; Online. DOI: 10.1038/srep31107