This project develops and validates new pediatric sleep monitoring devices and AI-based diagnostic tools. Traditional methods like polysomnography (PSG) are precise but resource-intensive, limiting large-scale or long-term monitoring. New wearable and simplified devices will be adapted for children, focusing on comfort, usability, and ability to capture sleep stages and sleep microstructure. Children and caregiver feedback will guide refinement of devices and clinical practices. Validation will compare new tools against gold-standard PSG in children with and without chronic disease from the CopenSleep cohort.
AI models will be trained on large heterogeneous datasets to automate sleep scoring and sleep microstructure detection. These models will improve precision, reduce inter-scorer variability, and allow scalable analysis of pediatric sleep data. Integration with national databases enables long-term outcome studies linking sleep to health, cognition, and wellbeing. Expected outcomes include validated child-friendly devices, robust AI methods, and improved diagnostics for sleep-wake disturbances.