EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Predicting Sleep and Sleep Stage in Children Using Actigraphy and Heartrate via a Long Short‐Term Memory Deep Learning Algorithm: A Performance Evaluation

R Glenn Weaver iD, James W White, Olivia Finnegan, Hongpeng Yang, Zifei Zhong, Keagan Kiely, Catherine Jones iD, Yan Tong, Srihari Nelakuditi, Rahul Ghosal, David E Brown, Russ Pate und 7 weitere

Journal of Sleep Research · 2026

DOI: 10.1111/jsr.70149 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Sleep Research, 2026. Verfasst von R Glenn Weaver und 18 weiteren Autor:innen. Beteiligte Einrichtung: Department of Exercise Science, Arnold School of Public Health, University of South Carolina, Columbia, South Carolina, USA. Hauptthemen laut MEDLINE: Actigraphy, Deep Learning, Sleep Stages, Sleep. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 49 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NIGMS NIH HHS, NIDDK NIH HHS, National Institute of Diabetes and Digestive and Kidney Diseases of the National Institutes of Health.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Children's ambulatory sleep is commonly measured via actigraphy. However, traditional actigraphy measured sleep (e.g., Sadeh algorithm) struggles to predict wake (i.e., specificity, values typically < 70) and cannot predict sleep stages. Long short‐term memory (LSTM) is a machine learning algorithm that may address these deficiencies. This study evaluated the agreement of LSTM sleep estimates from actigraphy and heartrate (HR) data with polysomnography (PSG). Children ( N = 238, 5–12 years, 52.8% male, 50% Black 31.9% White) participated in an overnight laboratory polysomnography. Participants were referred because of suspected sleep disruptions. Children wore an ActiGraph GT9X accelerometer and two of three consumer wearables (i.e., Apple Watch Series 7, Fitbit Sense, Garmin Vivoactive 4) on their non‐dominant wrist during the polysomnogram. LSTM estimated sleep versus wake and sleep stage (wake, not‐REM, REM) using raw actigraphy and HR data for each 30‐s epoch. Logistic regression and random forest were also estimated as a benchmark for performance with which to compare the LSTM results. A 10‐fold cross‐validation technique was employed, and confusion matrices were constructed. Sensitivity and specificity were calculated to assess the agreement between research‐grade and consumer wearables with the criterion polysomnography. For sleep versus wake classification, LSTM outperformed logistic regression and random forest with accuracy ranging from 94.1 to 95.1, sensitivity ranging from 94.9 to 95.9 across different devices, and specificity ranging from 84.5 to 89.6. The addition of HR improved the prediction of sleep stages but not binary sleep versus wake. LSTM is promising for predicting sleep and sleep staging from actigraphy data, and HR may improve sleep stage prediction.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

ActigraphyDeep LearningSleep StagesSleepHumansMaleChildFemalePolysomnographyChild, PreschoolAlgorithmsWearable Electronic DevicesLong Short Term Memory

Stichwörter der Autor:innen

LSTMambulatory sleepdevice agnosticmachine learningyouth

Autor:innen und Institutionen

  • Department of Exercise Science, Arnold School of Public Health, University of South Carolina, Columbia, South Carolina, USA.
  • Department of Computer Science and Engineering, University of South Carolina, Columbia, South Carolina, USA.
  • Kinesiology, College of Health and Human Sciences, Iowa State University, Ames, Iowa, USA.
  • Center for Health Sciences, SRI, Menlo Park, California, USA.

ORCID-Profile: R Glenn Weaver, Catherine Jones, Sarah Burkart

Förderung

  • NIGMS NIH HHS
  • NIDDK NIH HHS
  • National Institute of Diabetes and Digestive and Kidney Diseases of the National Institutes of Health
  • NICHD NIH HHS
  • National Institute of General Medical Sciences

Zitieren

APA

Weaver, R. G., White, J. W., Finnegan, O., Yang, H., Zhong, Z., Kiely, K., Jones, C., Tong, Y., Nelakuditi, S., Ghosal, R., Brown, D. E., Pate, R., Welk, G. J., de Zambotti, M., Wang, Y., Burkart, S., Adams, E. L., Armstrong, B., & Beets, M. W. (2026). Predicting Sleep and Sleep Stage in Children Using Actigraphy and Heartrate via a Long Short‐Term Memory Deep Learning Algorithm: A Performance Evaluation. Journal of Sleep Research. https://doi.org/10.1111/jsr.70149

Vancouver

Weaver RG, White JW, Finnegan O, Yang H, Zhong Z, Kiely K, et al. Predicting Sleep and Sleep Stage in Children Using Actigraphy and Heartrate via a Long Short‐Term Memory Deep Learning Algorithm: A Performance Evaluation. Journal of Sleep Research. 2026. doi:10.1111/jsr.70149

BibTeX
@article{Weaver2026,
  author = {Weaver, R Glenn and White, James W and Finnegan, Olivia and Yang, Hongpeng and Zhong, Zifei and Kiely, Keagan and Jones, Catherine and Tong, Yan and Nelakuditi, Srihari and Ghosal, Rahul and Brown, David E and Pate, Russ and Welk, Gregory J and de Zambotti, Massimiliano and Wang, Yuan and Burkart, Sarah and Adams, Elizabeth L and Armstrong, Bridget and Beets, Michael W},
  title = {Predicting Sleep and Sleep Stage in Children Using Actigraphy and Heartrate via a Long Short‐Term Memory Deep Learning Algorithm: A Performance Evaluation},
  journal = {Journal of Sleep Research},
  year = {2026},
  doi = {10.1111/jsr.70149}
}
Kostenlos starten →