Related Experiment Video
Updated: Aug 7, 2026
07:40
Multi-Modal Home Sleep Monitoring in Older Adults
Published on: January 26, 2019
The Human Sleep Project: A Multi-Center Clinical Polysomnography Dataset Across the Human Lifespan
Qichen Li1, Shenghan Wen1, Haoqi Sun2,3
1Emory University School of Medicine, Atlanta, GA, USA.
Sleep
|August 5, 2026
Summary
The Human Sleep Project created a large, multi-center clinical polysomnography (PSG) dataset to advance sleep medicine research. This resource enables scalable, reproducible studies across diverse populations and age groups.
Area of Science:
- Sleep Medicine
- Clinical Neuroscience
- Big Data in Healthcare
Background:
- Existing polysomnography (PSG) datasets are limited in size, diversity, and generalizability.
- Single-center, age-specific, or epidemiological data restrict large-scale clinical research in sleep.
- A need exists for a comprehensive, multi-center clinical PSG dataset.
Purpose of the Study:
- To establish a large, multi-center clinical PSG dataset.
- To support scalable and broadly representative research in sleep medicine and clinical neuroscience.
- To overcome limitations of existing sleep research resources.
Main Methods:
- Collected 119,234 overnight sleep recordings from over 90,000 patients across five major US institutions.
- Included standard PSG montage, clinical, and automated sleep annotations.
- Harmonized data using Brain Imaging Data Structure (BIDS) and linked to de-identified EHR data.
Main Results:
- The dataset spans the full human lifespan and includes 22 ICD-10 clinical categories.
- Confirmed age-related trends in sleep architecture (e.g., reduced N3 sleep, increased fragmentation).
- Standardization successfully addressed inter-site variability, ensuring data consistency.
Conclusions:
- The Human Sleep Project provides harmonized data for 119,234 sleep studies.
- Enables reproducible sleep research at an unprecedented scale.
- Facilitates advanced analyses of sleep physiology, disorders, and neuroscience.
