Visualizing the Motor Movements of Autism Using Motion Capture Animation to Facilitate Differential Diagnosis



Master's Research Project

Committee members:

Shelley Wall, AOCAD, MScBMC, PhD
Biomedical Communications
Institute of Medical Science, Faculty of Medicine
University of Toronto Mississauga

Marc Dryer, Hons BA, MSc, MScBMC
Biomedical Communications
Institute of Medical Science, Faculty of Medicine
University of Toronto Mississauga

Kassia Johnson, MD, FRCPC
Developmental Paediatrics
Surrey Place Centre
Toronto




Back to portfolio




Abstract



Autism Spectrum Disorders (ASD) are spectrum-based neurodevelopmental disorders which demonstrate specific motor impairments. Much research has gone into characterizing the differences in movements between autism and other diagnostic groupings of neurodevelopmental disorders. However, current techniques for studying movement, which involve video analyses and direct observation, remain limited because they do not allow researchers to view movements in a controlled and consistent manner that can facilitate comparisons over large sample populations. This project will be the first to use motion capture (MoCap) animation to study the movements of ASD in an attempt to address this concern.



Introduction



Autism Spectrum Disorders are spectrum-based neurodevelopmental disorders comprised of Autistic Disorder, Asperger's Syndrome, and “Pervasive Developmental Disorder - not otherwise specified”. The prevalence is estimated at 1 in 150 children and is three to four times more common in boys than girls. Diagnosis is based on behavioural criteria outlined in the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders, which include impaired ability to engage in social interaction, impaired communication skills, and specific characteristic behavioural patterns (e.g. preoccupation, resistance to change, adherence to nonfunctional routines, stereotyped and repetitive behaviours). To date, there is no known cure for the condition and its aetiology remains largely unknown.

Motor impairments are frequently associated with ASD. These impairments often involve clumsiness and difficulties in accurately completing gross and fine motor tasks. Many individuals with ASD demonstrate stereotyped and repetitive movements. These movements can seem to follow elaborate patterns used to stimulate the individual’s sensory systems or appear purposeless but may in fact be a part of the individual’s attempts to calm, motivate, or distract themselves. These “stereotypies” include hand flapping, spinning, pacing, facial grimacing, and rocking of the body. The real motivation or reason for engaging in these behaviours is poorly understood.

Much work has gone into characterizing the differences in stereotyped movements between autism and other diagnostic groupings of neurodevelopmental disorders. However, current research remains limited for two reasons. The first is that stereotyped movements are difficult to analyze and compare using video analyses and direct observation. This leads directly to the second reason – the lack of a reliable metric with which to compare movements. There is a need for techniques that allow researchers to view movements in a controlled and consistent manner that can facilitate comparisons over large sample populations.

This project will be the first to use MoCap animation to study the stereotyped movements of ASD and will act as a proof of concept to address the limitations outlined above. A software application capable of presenting interactive MoCap animations of individuals with ASD and other neurodevelopmental disorders will be created to give researchers a tool to analyze motor movements in a way that has never been explored.



Objectives



Specific neurobehavioural research objectives include:


1) Assessing the potential use of MoCap animation to study motor movements in neurodevelopmental disorders, in particular, ASD

2) Identifying the differences in motor movements in particular stereotyped movements between four groups – children with ASD, intellectual disability with autism, intellectual disability without autism, and typically developed children


Visual research objectives include:


1) Determining the most effective type of three-dimensional (3D) character model to animate in order to visualize differences in the motor movements of individuals with neurodevelopmental disorders, in particular, ASD

2) Exploring whether camera interactivity when viewing an animation can aid in the understanding and perception of movement



Methods



The animation viewer will be created using the Unity 3D game development tool. A game engine like Unity allows for 3D scenes to be rendered in real-time, giving users the ability to dynamically interact with animations while viewing them. The main challenge of this project will be to come up with a visualization technique that can capture the subtle complexities of biological motion while facilitating optimized comparisons between movements. As such, this project will have a strong information visualization component and a variety of techniques will need to be explored (e.g. motion trails, dynamically generated motion charts). The success and effectiveness of the interface design, character models, interactivity, and information visualization techniques will be evaluated based on user feedback from the research committee and staff at Surrey Place Centre, an interdisciplinary community-based agency in Toronto that offers clinical services and programs to individuals with developmental disabilities.



Conclusion



The results of this study will provide insight into if and how motor movement variations between ASD and other classes of neurodevelopmental disorders can be better characterized. As well, it will explore the viability of using MoCap animation in a diagnostic context, potentially opening a new avenue of study where computer animation can be used to visualize motor movements for differential diagnosis.



Back to portfolio | Back to top