Can Sleep Architecture and Behavior Patterns Predict Next-Day Challenging Behavior in Autism?

Abstract

This presentation discusses our ongoing research into the relationship between sleep quality and challenging behaviors in children with autism spectrum disorder. The talk highlights our novel methodological approach using non-intrusive sensing technology and privacy-preserving edge computing to monitor sleep patterns without disrupting the home environment. We present preliminary findings from our collaboration with The Center for Discovery, demonstrating how specific sleep architecture disruptions may serve as predictive biomarkers for next-day behavioral challenges, offering potential pathways for early intervention.

Key Points

• Development and validation of off-body sleep monitoring systems
• Privacy-preserving computation approaches for vulnerable populations
• Correlation between specific sleep parameters and behavioral outcomes
• Application of machine learning for behavioral prediction
• Implementation challenges in real-world settings
• Translation into clinical practice and personalized intervention strategies

Acknowledgments

This research is supported by the Thrasher Research Fund Early Career Award and conducted in collaboration with The Center for Discovery.