ANPHY-Sleep: Open Sleep Database from Healthy Adults with High-Density EEG
Sleep is a fundamental process that plays a crucial role in various physiological functions, including memory consolidation, synaptic plasticity, and brain metabolic waste clearance. Disruption of the intricate relationship between sleep and other physiological processes can have significant medical consequences. Therefore, studying the brain activity of healthy individuals during sleep serves as a valuable resource for understanding normal brain function and identifying abnormal activity associated with neurological and psychiatric disorders.
Polysomnography (PSG) is the gold standard for measuring physiological sleep, involving multiple electrodes to record neural events, eye movements, muscle activity, and other parameters during sleep. While there are some open PSG sleep datasets available, they often have limitations in terms of electrode density and sampling frequency, restricting the exploration of spatial characteristics of sleep.
The ANPHY-Sleep database presents a unique dataset comprising high-density EEG (HD-EEG) sleep recordings from 29 healthy subjects. These recordings include 83 channels, along with EOG, EMG, ECG, and detailed sleep scoring annotations. The dataset also provides information on sleep demographics, macrostructure, and spectral density in different sleep stages and regions, both at individual and group levels.
The dataset offers researchers a valuable resource for studying healthy human sleep at a high spatio-temporal resolution. It can be used for developing automatic algorithms for sleep feature detection, such as sleep stage classification, and for comparing sleep patterns between healthy individuals and those with sleep disorders.
The experiment design involved screening healthy adults between 18 and 45 years, installing HD-EEG and PSG setups, collecting overnight sleep recordings, scoring sleep stages, preprocessing data to remove artifacts, and validating the dataset for public sharing. The dataset’s technical validation included demographic analysis, sleep macrostructure features, and spectrum analysis in different brain regions during various sleep stages.
The ANPHY-Sleep database is publicly available on the Open Science Framework (OSF) and includes individual folders for each study participant, containing overnight sleep recordings, sleep annotations, artifact detection matrices, and detailed demographic information. The dataset structure and specifications are outlined, providing researchers with a comprehensive resource for studying healthy sleep patterns.
In conclusion, the ANPHY-Sleep database offers researchers a unique opportunity to explore healthy human sleep patterns at a high resolution, develop automated sleep analysis algorithms, and compare sleep features between healthy individuals and those with sleep disorders. By leveraging this open-access dataset, researchers can advance our understanding of sleep physiology and pathology, leading to improved diagnosis and treatment of sleep-related conditions.