The Bitbrain Open Access Sleep (BOAS) dataset

Source
DataCite
Published
2026
License
Creative Commons Zero v1.0 Universal
Themes
Public Health, Machine Learning & AI

Get the data from DataCite

About this dataset

The Bitbrain Open Access Sleep (BOAS) dataset comprises 128 nights of simultaneous polysomnography and wearable EEG recordings from 108 unique participants, designed to validate emerging wearable EEG technology against clinical-grade sleep monitoring standards. The dataset includes human expert consensus sleep stage labels derived from three independent scorers following AASM criteria, as well as AI-generated sleep stage labels from a deep learning model trained on both PSG and wearable EEG data. This dual-recording approach provides a comprehensive resource for evaluating wearable EEG performance in sleep studies and advancing automated sleep staging algorithms.

How the data was collected

The dataset includes human expert consensus sleep stage labels derived from three independent scorers following AASM criteria, as well as AI-generated sleep stage labels from a deep learning model trained on both PSG and wearable EEG data.

Browse sustainability datasets on SusanHub