According to researchers, the new findings may make it easier to identify cases of autism in children by analyzing short videos.
“We recognize that while the diagnosis process understands the indexing of the specific behaviors, ASD also comes with broad impairments that often transcend single behavioral acts,” the authors explained in their findings.
“For instance, the atypical nonverbal behaviors manifest through global patterns of atypical postures and movements, fewer gestures used and often decoupled from visual contact, facial affect, speech.”
“Here, we tested the hypothesis that a deep neural network trained on the non-verbal aspects of social interaction can effectively differentiate between children with ASD and their typically developing peers,” the authors also explained.
The study included 68 children diagnosed with autism and 68 without any neurodevelopmental condition.
The model used to identify autism peaked at an accuracy rate of nearly 80 percent, indicating effectiveness in classifying the vast majority of cases of autism in young children by analyzing short videos.
“Our model achieves an accuracy of 80.9% with the prediction probability positively correlated to the overall level of symptoms of autism in social affect and repetitive and restricted behaviors domain,” the authors concluded. “Provided the non-invasive and affordable nature of computer vision, our approach carries reasonable promises that a reliable machine-learning-based ASD screening may become a reality not too far in the future.”