PaperarXiv:2412.15947

Mamba-based Deep Learning Approach for Sleep Staging on a Wireless Multimodal Wearable System without Electroencephalography

Venue listed2026Added Sep 26, 2026

Authors: Andrew H. Zhang, Alex He-Mo, Richard Fei Yin, Chunlin Li, Yuzhi Tang, Dharmendra Gurve, Veronique van der Horst, Aron S. Buchman et al. (12 authors)

SLEEP (arXiv q-bio.QM / cs.LG)

PDF arXiv2412.15947 Publisher Cite

TL;DR

Mamba-based RNN ensemble for sleep staging from Sibel ANNE One dual-module wearable (chest ECG+ACC+temp; finger PPG+temp) without EEG, n=357 adults with concurrent tertiary-care PSG. After ensembling: 3-class κ ~0.73; 4-class κ ~0.62; 5-class κ ~0.53 (balanced accuracies 84.0% / 75.3% / 65.1%). Demonstrates multimodal wearable (incl.

Form factor: finger

Abstract

Study Objectives: We investigate a Mamba-based deep learning approach for sleep staging on signals from ANNE One (Sibel Health, Evanston, IL), a non-intrusive dual-module wireless wearable system measuring chest electrocardiography (ECG), triaxial accelerometry, and chest temperature, and finger photoplethysmography and finger temperature. Methods: We obtained wearable sensor recordings from 357 adults undergoing concurrent polysomnography (PSG) at a tertiary care sleep lab. Each PSG recording was manually scored and these annotations served as ground truth labels for training and evaluation of our models. PSG and wearable sensor data were automatically aligned using their ECG channels with manual confirmation by visual inspection. We trained a Mamba-based recurrent neural network architecture on these recordings. Ensembling of model variants with similar architectures was performed. Results: After ensembling, the model attains a 3-class (wake, non rapid eye movement [NREM] sleep, rapid eye movement [REM] sleep) balanced accuracy of 84.02%, F1 score of 84.23%, Cohen's of 72.89%, and a Matthews correlation coefficient (MCC) score of 73.00%; a 4-class (wake, light NREM [N1/N2], deep NREM [N3], REM) balanced accuracy of 75.30%, F1 score of 74.10%, Cohen's of 61.51%, and MCC score of 61.95%; a 5-class (wake, N1, N2, N3, REM) balanced accuracy of 65.11%, F1 score of 66.15%, Cohen's of 53.23%, MCC score of 54.38%. Conclusions: Our Mamba-based deep learning model can successfully infer major sleep stages from the ANNE One, a wearable system without electroencephalography (EEG), and can be applied to data from adults attending a tertiary care sleep clinic.

Verbatim from arXiv (opens in a new tab) · arXiv metadata (abstracts are released under CC0)

Results on ChatPPG leaderboards

All leaderboards

No results from this paper on ChatPPG leaderboards yet. Browse the leaderboards.

Tasks & signals

Tasks

Sleep staging

Signals

PPG ECG IMU Temp

Keywords

sleepstagingMambawearablefinger-PPGno-EEG

How it works

Cite this paper

APA

Zhang, A. H., He-Mo, A., Yin, R. F., Li, C., Tang, Y., Gurve, D., van der Horst, V., Buchman, A. S., et al. (2026). Mamba-based Deep Learning Approach for Sleep Staging on a Wireless Multimodal Wearable System without Electroencephalography. SLEEP. https://doi.org/10.1093/sleep/zsag022

BibTeX

@article{zhang2026mambabased,
  title         = {{Mamba-based Deep Learning Approach for Sleep Staging on a Wireless Multimodal Wearable System without Electroencephalography}},
  author        = {Zhang, Andrew H. and He-Mo, Alex and Yin, Richard Fei and Li, Chunlin and Tang, Yuzhi and Gurve, Dharmendra and van der Horst, Veronique and Buchman, Aron S. and others},
  year          = {2026},
  journal       = {SLEEP},
  doi           = {10.1093/sleep/zsag022},
  eprint        = {2412.15947},
  archivePrefix = {arXiv}
}

Built from the hub record. Check it against the publisher before submitting.

Sep 26, 2026Added to the ChatPPG research hub

Mamba-based Deep Learning Approach for Sleep Staging on a Wireless Multimodal Wearable System without Electroencephalography

TL;DR

Abstract

Results on ChatPPG leaderboards