An interdisciplinary study group ofUniversity of Geneva (UNIGE), Switzerland, has developed a artificial intelligence (AI) algorithm based onautomated video analysis, which allows to study the non-verbal communication of children in an anonymous and standardized way. Easy to use, this technology correctly classified 80% of cases from short videos showing a child with or without autism under the age of 5 playing with an adult.
Example of ‘skeleton’ extraction using OpenPose on an image extracted from a social scene recording: (A) OpenPose key points superimposed on the video recording (B). OpenPose key points on a black background, preserving the bare minimum of scene characteristics needed to identify social interaction. Credits: Kojovic, Natraj, Mohanty, Maillart, Schaer
Individuals with autism spectrum disorder often have communication problems and difficulties in social interactions. Although very common, this disorder is difficult to diagnose before the age of five. However, early care can compensate for these difficulties by providing specific behavioral interventions focused on developing the abilities affected by autism.
The results of the Research have been published in the scientific journal Scientific Reports.
Automated video analysis: This is how early detection of autism spectrum disorder works
Autism spectrum disorder involves a child 54 and is characterized by difficulties in social interactions, impaired communication skills and the presence of repetitive behaviors and restricted interests. Consequentially, children with autism often have difficulty following a standard school curriculum.