Automated Quantification of Stereotypical Motor Movements Occurring in Autism Spectrum Disorder

Background:  The prevalence of autism spectrum disorder (ASD) has risen significantly in the last ten years, and today roughly 1 in 68 children have been diagnosed. Stereotypical motor movements exhibited by patients with autism—which include spinning, body-rocking, and hand-flapping—tend to alienate people and render social interaction difficult. Despite the growing number of individuals affected by autism, an effective, accurate method of automatically quantifying such movements remains unavailable.

Objectives:  The focus of this research is to utilize Kinect v2, software developed by Microsoft to detect body movements, as a means of objectively and systematically tracking these movements.

Methods:  The Kinect camera was used to film 12 actors, each performing the three movements. MATLAB and Visual Gesture Builder (VGB), a program that generates data on gesture detection, were used to analyze the skeletal structures in these recordings, after which the validity and reliability of the data were tested by comparing the outputs to that of manual grading.

Results:  It was concluded that VGB analysis is a more accurate method of automatically quantifying the stereotypical motor movements than MATLAB.

Conclusions:  Our findings suggest that Kinect is a potentially useful tool for automatically quantifying motor symptoms in patients with autism.