Skip to content
DraconiticResearch
Search Sources API
194,161 studies31 sourcesUpdated 50 min ago
Peer-reviewedCC BYNPJ digital medicine

AnySleep: a channel-agnostic deep learning system for high-resolution sleep staging in multi-center cohorts.

  • Niklas Grieger1,2,3
  • Jannik Raskob4,5
  • Siamak Mehrkanoon6
  • Stephan Bialonski7,8
8 affiliations
  1. 1Department of Medical Engineering and Technomathematics, FH Aachen University of Applied Sciences, Jülich, Germany. [email protected].
  2. 2Department of Information and Computing Sciences, Utrecht University, Utrecht, The Netherlands. [email protected].
  3. 3Institute for Data-Driven Technologies, FH Aachen University of Applied Sciences, Jülich, Germany. [email protected].
  4. 4Department of Medical Engineering and Technomathematics, FH Aachen University of Applied Sciences, Jülich, Germany.
  5. 5Institute for Data-Driven Technologies, FH Aachen University of Applied Sciences, Jülich, Germany.
  6. 6Department of Information and Computing Sciences, Utrecht University, Utrecht, The Netherlands.
  7. 7Department of Medical Engineering and Technomathematics, FH Aachen University of Applied Sciences, Jülich, Germany. [email protected].
  8. 8Institute for Data-Driven Technologies, FH Aachen University of Applied Sciences, Jülich, Germany. [email protected].
Read full textPDFDOICite

Abstract

Sleep is essential for health, yet studying its dynamics requires manual sleep staging, a labor-intensive step in research and clinical care. Across centers, polysomnography (PSG) recordings are traditionally scored in 30-s epochs for pragmatic, not physiological, reasons and vary in electrode count, montage, and subject characteristics. These constraints challenge harmonized multi-center studies and the discovery of robust biomarkers on shorter timescales. We present AnySleep, a deep neural network that scores sleep from any electroencephalography (EEG) or electrooculography (EOG) data at adjustable temporal resolutions. We trained and validated the model on over 20,000 overnight recordings (>200,000 h of EEG and EOG) from 28 datasets across multiple clinics to promote robust generalization across sites. The model attains state-of-the-art performance and surpasses or equals established baselines at 30-s epochs. Performance improves with more channels, yet remains strong when EOG is absent, or only EOG or single EEG derivations (frontal, central, or occipital) are available. On sub-30-s timescales, the model captures short wake intrusions consistent with arousals and improves the prediction of pathophysiological conditions (obstructive sleep apnea, narcolepsy type 1, insomnia) over 30-s scoring. We make the model publicly available to facilitate large-scale studies with heterogeneous electrode setups and accelerate biomarker discovery in sleep.

Related

  1. A context-aware temporal modeling through unified multi-scale temporal encoding and hierarchical sequence learning for single-channel EEG sleep staging
    Amir Ali Vakili, Salar Jahanshiri, Armin Salimi-Badr
    PLOS OnePeer-reviewedCC BY 4.0Full textNeurosciencePhysiologyPLOS
  2. MarkerScout: A Disease-Agnostic Machine Learning Framework for Biomarker Prediction from Multi-Scale Mechanistic Models
    Robert Moore, Frank Agayie-Ntim, Lindsey B. Crawford +5
    arXivPreprintCC BY-SA 4.0Full textGeneral biologyarXiv
  3. Phylogeny-agnostic strain-level prediction of phage–host interactions from genomes using machine learning
    Avery J. C. Noonan, Lucas Moriniere, Edwin O. Rivera-López +8
    Nature MicrobiologyPeer-reviewedCC BY-NC-ND 4.0Full text1 citedEcologyPMC
  4. AREION: A Foundation Model for Atlas-Agnostic Representation Learning of Brain Connectome
    Z. Zhang, H. Li, Y. Zhang +3
    bioRxivPreprintCC BY-NC-NDFull textNeurosciencebioRxiv
  5. Improving the Usability of Donor Data from Registries Using an Extreme Learning Machine Approach to Upgrade Low-/Mid- to High-Resolution HLA Data
    Stavros Kepentzis, Theofanis Chatzistamatiou, Jason Digalakis +3
    GenesPeer-reviewedCC BYFull textGenetics & genomicsEurope PMC

Cite

BibTeXRIS
BibTeX
@article{grieger2026anysleep,
  title = {{AnySleep: a channel-agnostic deep learning system for high-resolution sleep staging in multi-center cohorts.}},
  author = {Niklas Grieger and Jannik Raskob and Siamak Mehrkanoon and Stephan Bialonski},
  journal = {NPJ digital medicine},
  year = {2026},
  month = oct,
  doi = {10.1038/s41746-026-03406-7},
  url = {https://europepmc.org/article/MED/42963354},
  note = {CC BY},
}
RIS
TY  - JOUR
TI  - AnySleep: a channel-agnostic deep learning system for high-resolution sleep staging in multi-center cohorts.
AU  - Niklas Grieger
AU  - Jannik Raskob
AU  - Siamak Mehrkanoon
AU  - Stephan Bialonski
PY  - 2026
DA  - 2026/10/30
JO  - NPJ digital medicine
DO  - 10.1038/s41746-026-03406-7
UR  - https://europepmc.org/article/MED/42963354
AB  - Sleep is essential for health, yet studying its dynamics requires manual sleep staging, a labor-intensive step in research and clinical care. Across centers, polysomnography (PSG) recordings are traditionally scored in 30-s epochs for pragmatic, not physiological, reasons and vary in electrode count, montage, and subject characteristics. These constraints challenge harmonized multi-center studies and the discovery of robust biomarkers on shorter timescales. We present AnySleep, a deep neural network that scores sleep from any electroencephalography (EEG) or electrooculography (EOG) data at adjustable temporal resolutions. We trained and validated the model on over 20,000 overnight recordings (>200,000 h of EEG and EOG) from 28 datasets across multiple clinics to promote robust generalization across sites. The model attains state-of-the-art performance and surpasses or equals established baselines at 30-s epochs. Performance improves with more channels, yet remains strong when EOG is absent, or only EOG or single EEG derivations (frontal, central, or occipital) are available. On sub-30-s timescales, the model captures short wake intrusions consistent with arousals and improves the prediction of pathophysiological conditions (obstructive sleep apnea, narcolepsy type 1, insomnia) over 30-s scoring. We make the model publicly available to facilitate large-scale studies with heterogeneous electrode setups and accelerate biomarker discovery in sleep.
SN  - 2398-6352
ER  -