On Sleep-Wake Disorder Identification from Polysomnographic Representations
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.
Compared: self-supervised PSG FMs SleepFM, OSF, Hypnos vs sleep-stage-supervised scorers U-Sleep and SleepTransformer (pretrained in the SLEEPYLAND protocol on ~220k h of expert-annotated PSG from 17 NSRR cohorts; EMG/ECG/RESP counterparts trained the same way). Downstream: linear probing and a missingness-aware coupled-Mamba state-space model; fixed patient-grouped 5-fold CV.
- Venue: arXiv 2610.17451 (v1 2026-10-12, cs.LG)
- Signals: EEG, EOG, EMG, ECG, respiration
- Form factor: clinical
Abstract
Self-supervised polysomnography foundation models (PSG FMs) aim to capture clinically useful information beyond sleep staging, yet are rarely compared with sleep-staging encoders pretrained at comparable scale and modality coverage. Using a common whole-night, missingness-aware state-space downstream modeling framework, we compared frozen representations from prominent large-scale, multi-cohort sleep-stage-supervised encoders and self-supervised PSG FMs for multilabel identification of six sleep-wake and circadian disorder groups in 7,906 PSGs from an unseen registry. Although PSG FMs outperformed an EEG-only, single-cohort supervised baseline, this ordering reversed when supervised pretraining approached comparable scale and modality coverage. Transfer learning from neural sleep scorers achieved higher overall performance while exhibiting greater downstream data efficiency. These findings underscore the need for scale- and modality-matched pretrained supervised baselines when assessing PSG FMs' downstream potential.
Verbatim fromarXiv· arXiv metadata (abstracts are released under CC0)
How it works
Cite
@article{psgfmvssleepscorerssleep2026,
title = {{On Sleep-Wake Disorder Identification from Polysomnographic Representations}},
author = {Alvise Dei Rossi and Luigi Fiorillo and Giuliana Monachino and Alessia Calzoni and Julia van der Meer and Markus H. Schmidt and Claudio L. A. Bassetti and Silvia Santini and Francesca Faraci},
year = {2026},
journal = {arXiv 2610.17451 (v1 2026-10-12, cs.LG)},
eprint = {2610.17451},
url = {https://arxiv.org/abs/2610.17451}
}History
Added to the ChatPPG research hub
More on ChatPPG
- A meta-learning method with reduced data requirements for training and updating deep learning models on wearable devices
This paper proposes a meta-learning method that cuts how much labelled data a new wearable needs bef
- Agentic AI-enabled discovery across large-scale sleep physiology (AI Sleep Co-Scientist)
AI Sleep Co-Scientist is a setup in which human scientists direct specialist AI agents for hypothesi
- AnyECG-Echo: Wearable Single-Lead ECG Detects Fine-Grained Structural Heart Disease Through Echo-Report Supervision
AnyECG-Echo uses contrastive pretraining between single-lead ECGs and echocardiography reports to sc