The following is a summary of a story that originally appeared on the Office of Interdisciplinary Programs .
While behavioral signs of autism can appear as early as six to 12 months, most children in the United States are not diagnosed until much later, with the average age around five years, according to Geraldine Dawson, founding director of the Duke Center for Autism and Brain Development.
An interdisciplinary team at Duke University hopes to change how autism is detected by leveraging technology and expertise from across campus.
Dawson partnered with engineers and computer scientists, launching a yearslong collaboration that combines clinical expertise with advanced technology. It led to the development of a digital screening tool, an app dubbed SenseToKnow, that parents can use on a smartphone or tablet.
"We know that toddlers who will go on to have a diagnosis of autism tend to pay more attention to non-social information than social information. So, we designed a video where the social information was on one side of the screen and the non-social on the other side," Dawson says.
Toddlers watch short videos while the device's camera records their responses. It captures eye gaze, facial expressions, and movement patterns, and then uploads the recordings to a secure server at Duke. By using AI-based computer vision tools to automatically analyze the detection of early behavioral signs, the tool aims to provide more accurate and accessible screening.
"I learned a lot because the computer is able to detect very subtle variations in behavior. The resolution and the precision are just so much greater than the human eye," Dawson says.
She and her team are now seeking clearance from the U.S. Food and Drug Administration. With guidance from the FDA, they are conducting a study of 200 children without an autism diagnosis and 150 with an autism diagnosis, recruited through Duke Primary Care.
"The SenseToKnow screening app is really one piece of a larger vision that we have for improving early access to care," Dawson says.
