Rosebrugh Bldg, Toronto, ON M5S 3G9
Room: RS 211
Autism spectrum disorder is a neurodevelopmental disorder defined by social communicationdifficulties and restricted/repetitive behaviours. Significant heterogeneity in genetic information,neurobiology, and behavioural characteristics has made it difficult to understand this disorder anddevelop appropriate interventions. To address this gap, we propose a machine learning basedapproach for characterizing the relationship between brain morphology and social behavioursymptomology. Participants will be 124 children and youth between the ages of 4 and 20 that havebeen clinically diagnosed with ASD. Participants are recruited via the Province of OntarioNeurodevelopmental Disorders Network (POND). Brain data will consist of corticometric,morphometric and volumetric measurements obtained from magnetic resonance imaging.Behavioural data will include severity scores obtained from the Social CommunicationQuestionnaire (SCQ) which probes the core domains involved in ASD. We will use supervisedand unsupervised learners to predict social communication symptom severity from brainmorphology. We anticipate 1) a non-linear relation between brain morphology and socialcommunication functions and 2) a subset of brain regions contributing to this relation. This studywill provide a better understanding of ASD by unraveling the effects of brain morphology on socialskills and highlighting significant brain regions contributing to social function.