Objective:

To develop an interpretable multi-agent multimodal framework for analyzing the complex interactions of factors influencing sleep disorders.

Approach:

• Framework Development: Created a framework with three specialized agents: Data Analyst Agent, Physiology and Health Analyst Agent, and Validation Analyst Agent.
• Data Analysis: Applied the framework to public and synthetic sleep-health datasets to identify correlations and interactions.

Key Findings:

• Weak or uncertain associations were found between bedtime consistency, light exposure, caffeine intake, stress, heart rate, and sleep outcomes.
• High caffeine intake was linked to increased sleep-disorder risk.
• Different types of physical activity showed varying effects on sleep disorders, with agility drills linked to insomnia and endurance running associated with sleep apnea.
• Occupational context, psychological stress, stimulant use, and physiological indicators collectively influenced sleep disorder profiles.

Interpretation:

The framework enhances interpretability and supports evidence-grounded reasoning in sleep analysis.

Limitations:

• The second dataset used was synthetic, limiting external clinical validation.
• Several pairwise physiological associations were weak.

Conclusion:

The study highlights the need for integrated analysis of sleep-related determinants.