Agentic AI-enabled discovery across large-scale sleep physiology

2026-07-27

AI Summary

This study addresses the challenges of uncovering disease associations in large-scale multimodal sleep physiology data, where strong reliance on expert knowledge and the limitations of general-purpose AI hinder progress. The authors propose the AI Sleep Co-Scientist system—a novel expert-guided multi-agent collaborative framework applied for the first time to ultra-large-scale sleep research—enabling an interpretable and reproducible automated scientific workflow across 124,000 polysomnography records. Integrating physiological signal processing, late-fusion sleep-age modeling, transient oscillation analysis, and survival analysis, the system reveals that reduced sleep network coupling significantly predicts Parkinson’s disease (HR=1.48) and Alzheimer’s disease (HR=1.38), constructs a structured sleep-age model outperforming early-fusion approaches, and uncovers intermediate phenotypes linking insomnia and sleep apnea, regulatory patterns of REM rhythm, and characteristic EEG abnormalities in narcolepsy type 1.

Abstract

Sleep occupies roughly one-third of human life, yet many aspects of its physiology remain poorly understood. Large polysomnography (PSG) datasets offer new opportunities to study sleep and its links to disease, but extracting insight from these recordings requires substantial expert effort and remains difficult for general-purpose AI systems. We developed AI Sleep Co-Scientist, an expert-guided environment in which human scientists direct specialist agents for hypothesis development, signal preprocessing, and statistical analysis, reviewing intermediate outputs. Each reported result is linked to the executable code that produced it. Across four cohorts of approximately 124,000 PSG recordings and more than 50 TB of raw signals, we conducted five case studies spanning how sleep physiology relates to future disease, how it distinguishes clinical phenotypes, and how sleep is organized and regulated. Diminished network-level physiological coupling during sleep was associated with incident Parkinson's disease (HR 1.48) and Alzheimer's disease (HR 1.38). A physiologically structured late-fusion sleep-age model outperformed an unconstrained early-fusion approach, and its age residual was associated with incident disease across multiple organ systems. Arousal dynamics characterized comorbid insomnia and sleep apnoea as an intermediate phenotype skewed towards obstructive sleep apnoea, distinguished by prolonged post-arousal wakefulness. Rapid eye movement (REM) bout duration tracked preceding non-REM sleep more closely than intervening wakefulness. Transient-oscillation analysis identified a fast-sigma deficit and excess centrofrontal theta activity in narcolepsy type 1. Together, these findings connect sleep to disease risk, clinical classification, and its own regulation, and show how agentic AI can support large-scale, multimodal discovery.

Problem

• sleep physiology
• large-scale data
• disease association
• clinical phenotyping
• physiological regulation

Innovation

• agentic AI
• polysomnography
• sleep physiology
• AI co-scientist
• multimodal discovery