EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Detection of sleep apnea using only inertial measurement unit signals from apple watch: a pilot-study with machine learning approach

Junichiro Hayano iD, Mine Adachi, Yutaka Murakami, Fumihiko Sasaki, Emi Yuda iD

Sleep and Breathing · 2025-03 · 29 (1)

DOI: 10.1007/s11325-025-03255-w ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sleep and Breathing, online erschienen am 01.02.2025. Verfasst von Junichiro Hayano und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Research and Development, Heart Beat Science Lab Inc., Nagoya, Japan. hayano@acm.org. Hauptthemen laut MEDLINE: Machine Learning, Polysomnography, Sleep Apnea Syndromes, Signal Processing, Computer-Assisted. Die Arbeit wurde bisher 15-mal zitiert (Crossref) und nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Sony Group Inc..

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Purpose

Despite increased awareness of sleep hygiene, over 80% of sleep apnea cases remain undiagnosed, underscoring the need for accessible screening methods. This study presents a method for detecting sleep apnea using data from the Apple Watch’s inertial measurement unit (IMU).

Methods

An algorithm was developed to extract seismocardiographic and respiratory signals from IMU data, analyzing features such as breathing and heart rate variability, respiratory dips, and body movements. In a cohort of 61 adults undergoing polysomnography, we analyzed 52,337 30-second epochs, with 12,373 (23.6%) identified as apnea/hypopnea episodes. Machine learning models using five classifiers (Logistic Regression, Random Forest, Gradient Boosting, k-Nearest Neighbors, and Multi-layer Perceptron) were trained on data from 41 subjects and validated on 20 subjects.

Results

The Random Forest classifier performed best in per-epoch respiratory event detection, achieving an AUC of 0.827 and an F1 score of 0.572 in the training group, and an AUC of 0.831 and an F1 score of 0.602 in the test group. The model’s per-subject predictions strongly correlated with the apnea-hypopnea index (AHI) from polysomnography ( r = 0.93) and identified subjects with AHI ≥ 15 with 100% sensitivity and 90% specificity.

Conclusion

Utilizing the widespread availability of the Apple Watch and the low power requirements of the IMU, this approach has the potential to significantly improve sleep apnea screening accessibility.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Machine LearningPolysomnographySleep Apnea SyndromesSignal Processing, Computer-AssistedHumansMaleFemalePilot ProjectsMiddle AgedAdultAlgorithmsAged

Stichwörter der Autor:innen

AccelerationApnea-hypopnea indexGyrocardiogramGyroscopeHome sleep apnea testingOut-of-center sleep testingSeismocardiogramWearable sensor

Autor:innen und Institutionen

  • Department of Research and Development, Heart Beat Science Lab Inc., Nagoya, Japan. hayano@acm.org.
  • Takaoka Clinic, Nagoya, Japan.
  • Sony Group Inc, Tokyo, Japan.
  • Department of Research and Development, Heart Beat Science Lab Inc., Nagoya, Japan.
  • Graduate School of Information Sciences, Tohoku University, Sendai, Japan.

ORCID-Profile: Junichiro Hayano, Emi Yuda

Förderung

  • Sony Group Inc.

Zitieren

APA

Hayano, J., Adachi, M., Murakami, Y., Sasaki, F., & Yuda, E. (2025). Detection of sleep apnea using only inertial measurement unit signals from apple watch: a pilot-study with machine learning approach. Sleep and Breathing, 29(1). https://doi.org/10.1007/s11325-025-03255-w

Vancouver

Hayano J, Adachi M, Murakami Y, Sasaki F, Yuda E. Detection of sleep apnea using only inertial measurement unit signals from apple watch: a pilot-study with machine learning approach. Sleep and Breathing. 2025;29(1). doi:10.1007/s11325-025-03255-w

BibTeX
@article{Hayano2025,
  author = {Hayano, Junichiro and Adachi, Mine and Murakami, Yutaka and Sasaki, Fumihiko and Yuda, Emi},
  title = {Detection of sleep apnea using only inertial measurement unit signals from apple watch: a pilot-study with machine learning approach},
  journal = {Sleep and Breathing},
  year = {2025},
  volume = {29},
  number = {1},
  doi = {10.1007/s11325-025-03255-w}
}
Kostenlos starten →