Autism, a developmental disorder that affects communication and behavior, is on the rise, with 1 in 59 children in the US being diagnosed with autism spectrum disorder (ASD), according to the CDC. However, early diagnosis and intervention can greatly improve outcomes for individuals with autism.
To advance early detection, researchers from the University of Southern California (USC) conducted a study using artificial intelligence and data from an iPad coloring game. The results, published in the Journal of Autism and Developmental Disorders, show promising potential for an easy and novel method of detecting autism.
The researchers used machine learning analytics to analyze touchscreen kinematic data collected during a five-minute coloring game played by 54 children between the ages of 8 and 17. By examining factors such as pressure and movement velocity, the AI was able to accurately distinguish between typically developing children and those with autism spectrum disorder with 76% accuracy.
Additionally, the study found that developmental coordination disorder, a motor skills disorder that often overlaps with autism, could also be distinguished from typical development with 78% accuracy, and from autism with 71% accuracy.
While the study was conducted on high-functioning children and adolescents and needs to be replicated in larger and more diverse groups, the researchers believe that early identification and tailored therapeutic approaches will lead to better long-term developmental outcomes for children with autism.
This study is just one example of the growing use of artificial intelligence in the field of autism detection and intervention. Researchers are also exploring the use of social robots to teach social and educational skills to children with autism. With ongoing advancements in technology, the hope is to identify and support children with autism as early as possible.
More Autism News:
