EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Early Drowsiness Detection via Second-Order Derivative Analysis of Heart Rate Variability: A Non-Contact ECG Approach with Machine Learning

Fabrice Vaussenat iD, Abhiroop Bhattacharya, Julie Payette iD, Alireza Saidi iD, Victor Bellemin, Geordi-Gabriel Renaud-Dumoulin, Sylvain G Cloutier, Ghyslain Gagnon iD

Sensors · 2026

DOI: 10.3390/s26041348 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, 2026. Verfasst von Fabrice Vaussenat und 7 weiteren Autor:innen. Beteiligte Einrichtung: Department of Electrical Engineering, École de Technologie supéRieure, Montreal, QC H3C 1K3, Canada. Hauptthemen laut MEDLINE: Heart Rate, Machine Learning, Electrocardiography, Sleep Stages. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 41 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: IRSST, NSERC.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Drowsy driving contributes to roughly 20% of traffic fatalities, yet most detection systems rely on behavioral cues that appear only after impairment has set in. Here we ask whether first and second derivatives of heart rate variability (HRV) can detect pre-crash states earlier than conventional approaches. Twenty-five participants completed 49 driving simulator sessions while we recorded cardiac activity through capacitive ECG electrodes embedded in the seat backrest—a non-contact method that avoids the privacy concerns of camera-based monitoring. To prevent circular evaluation, ground truth labels were based solely on crash proximity rather than HRV-derived scores. The combined HRV feature set (conventional metrics plus derivatives) achieved AUC = 0.863 for pre-crash prediction; derivatives alone reached only AUC = 0.573, indicating their value as complementary rather than standalone features. Driving performance indicators remained the strongest predictors (AUC = 0.999). Temporally, derivative-based detection preceded behavioral manifestations by 5–8 min and crash events by 6.8 ± 2.3 min. Across 1591 crashes and 6.78 million data points, we found that HRV derivatives capture physiological changes that precede overt impairment, though their utility depends on integration with other feature types.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Heart RateMachine LearningElectrocardiographySleep StagesHumansAutomobile DrivingAccidents, TrafficMaleFemaleAdult

Stichwörter der Autor:innen

HRV derivativescapacitive sensingdriver monitoring systemsdrowsy driving detectionheart rate variabilitymachine learningnon-contact ECGroad safety

Autor:innen und Institutionen

  • Department of Electrical Engineering, École de Technologie supéRieure, Montreal, QC H3C 1K3, Canada.
  • IRSST (Institut de Recherche Robert-Sauvé en Santé et en séCurité du Travail), 505 Blvd. De Maisonneuve Ouest, Montreal, QC H3A 3C2, Canada.

ORCID-Profile: Fabrice Vaussenat, Julie Payette, Alireza Saidi, Ghyslain Gagnon

Förderung

  • IRSST
  • NSERC

Zitieren

APA

Vaussenat, F., Bhattacharya, A., Payette, J., Saidi, A., Bellemin, V., Renaud-Dumoulin, G. G., Cloutier, S. G., & Gagnon, G. (2026). Early Drowsiness Detection via Second-Order Derivative Analysis of Heart Rate Variability: A Non-Contact ECG Approach with Machine Learning. Sensors. https://doi.org/10.3390/s26041348

Vancouver

Vaussenat F, Bhattacharya A, Payette J, Saidi A, Bellemin V, Renaud-Dumoulin GG, et al. Early Drowsiness Detection via Second-Order Derivative Analysis of Heart Rate Variability: A Non-Contact ECG Approach with Machine Learning. Sensors. 2026. doi:10.3390/s26041348

BibTeX
@article{Vaussenat2026,
  author = {Vaussenat, Fabrice and Bhattacharya, Abhiroop and Payette, Julie and Saidi, Alireza and Bellemin, Victor and Renaud-Dumoulin, Geordi-Gabriel and Cloutier, Sylvain G and Gagnon, Ghyslain},
  title = {Early Drowsiness Detection via Second-Order Derivative Analysis of Heart Rate Variability: A Non-Contact ECG Approach with Machine Learning},
  journal = {Sensors},
  year = {2026},
  doi = {10.3390/s26041348}
}
Kostenlos starten →