EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Deep Learning-Based Heartbeat Detection from 3D Seismocardiography for Robust Heart Rate Monitoring

Sobuz Rana iD, Jukka A Lipponen iD, Mika P Tarvainen iD

Sensors · 20.05.2026 · 26 (10) · 3254

DOI: 10.3390/s26103254 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 20.05.2026. Verfasst von Sobuz Rana und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Technical Physics, University of Eastern Finland, 70211 Kuopio, Finland. Hauptthemen laut MEDLINE: Deep Learning, Heart Rate. Die Arbeit nennt 44 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministry of Education and Culture.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate monitoring of heart rate (HR) is critical for assessing cardiac functions in a wide range of health and wellness applications. Seismocardiography (SCG), which captures subtle chest vibrations using wearable accelerometers, provides a non-invasive and cost-effective approach for resting and nocturnal HR monitoring. This study presents a deep learning-based approach for accurate heartbeat detection and HR estimation from three-dimensional SCG signals. The model was trained on a large-scale dataset of resting SCG signals collected from 6600 subjects and evaluated on an independent cohort of 947 individuals. For short-term (≤5 min) resting SCG recordings, the model achieved robust performance in heartbeat detection (PPV: 0.979, sensitivity: 0.916, F1-score: 0.946). HR estimation showed high accuracy, with a mean absolute error (MAE) of 0.27 bpm, root mean square error (RMSE) of 1.02 bpm, and correlation of 0.996 with the reference HR. To assess real-world applicability, the model was further evaluated on 28 nocturnal recordings acquired using Apple Watch accelerometer, yielding an MAE of 1.10 bpm, an RMSE of 1.88 bpm, and a correlation of 0.982. The proposed SCG-based deep learning model demonstrates robust and highly accurate HR monitoring in both resting and nocturnal conditions, highlighting its potential for integration with consumer-grade wearable devices in a server-based analysis pipeline.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningHeart RateHumansSignal Processing, Computer-AssistedMonitoring, PhysiologicWearable Electronic DevicesAccelerometry

Stichwörter der Autor:innen

accelerometersbeat detectiondeep learningheart rate monitoringseismocardiography

Autor:innen und Institutionen

  • Department of Technical Physics, University of Eastern Finland, 70211 Kuopio, Finland.

ORCID-Profile: Sobuz Rana, Jukka A Lipponen, Mika P Tarvainen

Förderung

  • Ministry of Education and Culture

Zitieren

APA

Rana, S., Lipponen, J. A., & Tarvainen, M. P. (2026). Deep Learning-Based Heartbeat Detection from 3D Seismocardiography for Robust Heart Rate Monitoring. Sensors, 26(10), 3254. https://doi.org/10.3390/s26103254

Vancouver

Rana S, Lipponen JA, Tarvainen MP. Deep Learning-Based Heartbeat Detection from 3D Seismocardiography for Robust Heart Rate Monitoring. Sensors. 2026;26(10):3254. doi:10.3390/s26103254

BibTeX
@article{Rana2026,
  author = {Rana, Sobuz and Lipponen, Jukka A and Tarvainen, Mika P},
  title = {Deep Learning-Based Heartbeat Detection from 3D Seismocardiography for Robust Heart Rate Monitoring},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {10},
  pages = {3254},
  doi = {10.3390/s26103254}
}
Kostenlos starten →