ChatPPG.
Paper

Pretraining on Sleep Data Improves non-Sleep Biosignal Tasks

William Lehn-Schiøler, Magnus Ruud Kjær, Phillip Hempel, Magnus Guldberg Pedersen, Rahul Thapa, Bryan He +4 more
arXiv preprint2026EEGECGPSG
arXiv Cite
PreprintFirst posted: May 4, 2026Added to atlas: Oct 11, 2026Source
Citations—Not available
GitHub stars—Not available
HF downloads—Not available

Activity counts measure attention, not model quality. Missing counts are unknown. Publication status follows the linked source; a listed venue alone does not confirm peer review.

This paper asks whether pretraining on sleep recordings helps with biosignal tasks outside sleep. The authors pretrained only on multimodal sleep data with a leave-one-out contrastive objective and tested on non-sleep EEG and ECG. Across eight tasks, sleep pretraining consistently beat training from scratch.

  • Pretraining: sleep-only multimodal contrastive (leave-one-out)
  • Evaluation: 8 non-sleep EEG and ECG tasks
  • Result: consistent gains over training from scratch
At a glance
Venue
arXiv preprint
Published
2026
DOI
Not reported
arXiv
2605.02500 (opens in a new tab)
PMID
Not reported
Licence
Not reported
Signals
EEGECGPSG
Form factor
Not reported
Subjects
Not reported
Code
Not reported
Added
Oct 11, 2026

Abstract

Sleep foundation models have recently demonstrated strong performance on in-domain polysomnography tasks, including sleep staging, apnea detection, and disease risk prediction. In this work, we investigate whether sleep biosignals can serve as an effective pretraining distribution for learning representations that transfer beyond sleep to adjacent domains. Following sleep foundation models, we perform sleep-only multimodal contrastive pretraining (with a leave-one-out objective) and evaluate transfer to non-sleep EEG and ECG, two well-benchmarked biosignal modalities with heterogeneous datasets and clinically meaningful downstream tasks. Across eight downstream tasks spanning multiple EEG and ECG datasets, sleep pretraining consistently improves performance relative to training from scratch. Moreover, on several tasks, we achieve performance competitive with or surpassing prior specialized state-of-the-art and foundation models.

Verbatim fromarXiv· arXiv metadata (abstracts are released under CC0)

Pretraining on Sleep Data Improves non-Sleep Biosignal Tasks: ChatPPG research summary card

How it works

How Method works: Setup → Method → ResultSetup: EEG (Electrophysiology); ECG (Electrocardiogram); PSG (Electrophysiology). Method: Method (Contrastive learning). Result: Evaluation (8 non-sleep EEG and ECG tasks); Result (consistent gains over training from scratch).SetupEEG: ElectrophysiologyEEGElectrophysiologyECG: ElectrocardiogramECGElectrocardiogramPSG: ElectrophysiologyPSGElectrophysiologyMethodMethodContrastive learningResultEvaluation: 8 non-sleep EEG and ECG tasksEvaluation8 non-sleep EEG and ECG tasksResult: consistent gains over training from scratchResultconsistent gains over training from…

Cite

@article{pretrainingonsleepdataim2026,
  title = {{Pretraining on Sleep Data Improves non-Sleep Biosignal Tasks}},
  author = {William Lehn-Schiøler and Magnus Ruud Kjær and Phillip Hempel and Magnus Guldberg Pedersen and Rahul Thapa and Bryan He and Nicolai Spicher and Andreas Brink-Kjaer and Lars Kai Hansen and Emmanuel Mignot},
  year = {2026},
  journal = {arXiv preprint},
  eprint = {2605.02500}
}

History

  1. Added to the ChatPPG research hub

More on ChatPPG