A team led by the Georgia Institute of Technology has received a $10 million "Expeditions in Computing" award from the National Science Foundation (NSF) to develop novel computing techniques for measuring and analyzing the behavior of children.
These technologies will be used to enable new approaches for identifying children at risk for autism and other developmental delays. In addition, these methods may potentially improve the delivery and evaluation of treatment.
The award -- one of only 10 given out by the NSF since 2008 -- provides up to $2 million in funding each year for five years and is designed to push boundaries in computer science. This project will push the limits by catalyzing a new scientific discipline called computational behavioral science, which will draw equally from computer science and psychology to transform the study of human behavior.
"There is a great deal of creativity in the computer science research community today," said Deborah Crawford, acting assistant director of Computer and Information Science and Engineering at NSF. "Our intentions with the Expeditions in Computing program are to stimulate and use that creativity to expand the horizons of computing. For example, several of the projects will be exploring new computational approaches to some of the most vexing problems we face in the science and engineering enterprise as well as in the larger society."
Autism affects one of every 110 children in the United States and the long-term outcomes for a child who is at risk for autism can be significantly improved if the child is treated at an early age. As a result, it is widely accepted that all children should be screened for developmental delays as early in life as possible.
"Direct observation of a child by highly trained specialists is an important step in assessing risk for developmental disorders, but such an approach cannot be easily scaled to the large number of individuals needing evaluation and treatment," said the project's lead principal investigator James Rehg, a professor in Georgia Tech's School of Interactive Computing.
For this project, the researchers will design vision, speech and wearable sensor technologies to analyze child behavior. Data will be collected from interactions between caregivers and children, children playing and socializing in a daycare environment, and clinicians interacting with children during individual therapy sessions. Multiple sensing technologies are necessary to obtain a comprehensive and integrated portrait of expressed behavior.
"People use eye gaze, hand gestures, facial expressions, and tone of voice to convey engagement and regulate social interactions," said co-principal investigator Gregory Abowd, a professor in the School of Interactive Computing at Georgia Tech. "In addition, physiological responses, such as increased heart rate, can impact the expression of these behaviors."
Cameras and microphones will provide an inexpensive and noninvasive way to measure eye gaze and facial and body expressions, along with speech and non-speech utterances. Wearable sensors will measure physiological variables such as heart rate and skin conductivity, which contain important clues about levels of internal stress and arousal that are linked to behavior.
The research team will also develop capabilities for synchronizing the signals from the microphones, cameras and on-body sensors. By developing and using models of social interactions, the researchers will analyze the sensor data to quantify engagement.
As part of this award, the researchers will use a behavioral screening instrument called Rapid-ABC, which is currently under development by Abowd, Emory University School of Medicine assistant professor of psychiatry Opal Ousley, and Georgia Tech School of Interactive Computing senior research scientist Rosa Arriaga. The researchers intend to utilize the information gathered from the microphones, cameras and on-body sensors to automate some of the scoring for the Rapid-ABC test.
"We hope that by incorporating this screening protocol into well-child doctor visits for children less than two years old, we can reduce the average age of autism diagnosis, which is currently about four years old," explained Arriaga.
In the future, the researchers hope to expand their work beyond autism to other developmental disorders and the general study of child behavior.
"While autism is our focus right now, this project addresses general social, communicative and repetitive behaviors, so the technologies we develop will have applicability to other childhood disorders, such as Down syndrome or attention deficit hyperactivity disorder," added Rehg.
Standardized screening leads to earlier autism diagnoses in pediatric care