A Personalized Classification Of Behavioral Severity Of Autism Spectrum Disorder Using A Comprehensive Machine Learning Framework

This study examines Autism Spectrum Disorder (ASD) by breaking it down into its behavioral components, using magnetic resonance imaging (MRI) scans from the publicly available dataset ABIDE II. Artificial intelligence (AI) models are used to select features and classify ASD severity and behavioral groups. The AI-based model achieved an average accuracy of 96%, and was able to accurately differentiate between the functionalities of specific brain regions. This AI-based model has the potential to accurately dissect ASD into its behavioral components.