The Role of Artificial Intelligence in Enhancing Identification of Autism in Children Through Motor Abnormalities

The research reported in this paper provides a systematic validation and illustrations of the potential of the non-invasive, sensor-less, Kinect-based temporal gait signal data in objective quantification and widely applicable objective identification of motor abnormalities in children with autism, a multidimensional neurodevelopmental disorder. This indicator, when verified and validated through more extensive work with a larger and representative sample, has the potential for an innovative and widely applicable objective identification tool to be utilized early on in routine pediatric and family practice for effective and timely referral for further comprehensive clinical and developmental evaluations from developmental teenage or older pediatric care to adulthood. Our findings and results lay the foundation for the next steps and development of predictive measurements and innovative optimal treatment personalization informed by creating innovative age group models that can enhance the everyday life experiences and long-term welfare of individuals with autism throughout various childhood stages, and for further enforceable legal and ethical person-protected international global appropriate use. The demands on healthcare and social services for young and older children with autism spectrum disorder (ASD), a multidimensional neurodevelopmental disorder, are continually increasing and pose substantial challenges for the family, school, medical, and social services. Community-based timely detection and identification of the potential treatment-elicited motor abnormalities, using a non-invasive technology that needs no physical dependency on the participants and trained staff who conduct the optimal data processing, is essential to further inform a timely, effective, and individualized behavioral interventions program pursued collaboratively by the family, early intervention specialists, and the child.