EUVIMEDEUVIMED entdecken
VergleichsstudieOpen Access · CC BY 4.0

Detecting Diverse Seizure Types with Wrist-Worn Wearable Devices: A Comparison of Machine Learning Approaches

Louis Faust iD, Jie Cui, Camille Knepper, Mona Nasseri, Gregory Worrell, Benjamin H Brinkmann iD

Sensors · 06.09.2025 · 25 (17) · 5562

DOI: 10.3390/s25175562 ↗

Kurz gefasst

Vergleichsstudie in der Zeitschrift Sensors, online erschienen am 06.09.2025. Verfasst von Louis Faust und 5 weiteren Autor:innen. Beteiligte Einrichtung: Robert D. and Patricia E. Kern Center for the Science of Health Care Delivery, Mayo Clinic, Rochester, MN 55905, USA. Hauptthemen laut MEDLINE: Wearable Electronic Devices, Seizures, Machine Learning. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 45 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NINDS NIH HHS, NCATS NIH HHS, the Robert D. and Patricia E. Kern Center for the Science of Health Care Delivery, the Mayo Neurology AI Program, the Mayo Center for Clinical and Translationa….

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objective

To evaluate the feasibility and effectiveness of wrist-worn wearable devices combined with machine learning (ML) approaches for detecting a diverse array of seizure types beyond generalized tonic–clonic (GTC), including focal, generalized, and subclinical seizures. Materials and Methods: Twenty-eight patients undergoing inpatient video-EEG monitoring at Mayo Clinic were concurrently monitored using Empatica E4 wrist-worn devices. These devices captured accelerometry, blood volume pulse, electrodermal activity, skin temperature, and heart rate. Seizures were annotated by neurologists. The data were preprocessed to experiment with various segment lengths (10 s and 60 s) and multiple feature sets. Three ML strategies, XGBoost, deep learning models (LSTM, CNN, Transformer), and ROCKET, were evaluated using leave-one-patient-out cross-validation. Performance was assessed using area under the receiver operating characteristic curve (AUROC), seizure-wise recall (SW-Recall), and false alarms per hour (FA/h).

Results

Detection performance varied by seizure type and model. GTC seizures were detected most reliably (AUROC = 0.86, SW-Recall = 0.81, FA/h = 3.03). Hyperkinetic and tonic seizures showed high SW-Recall but also high FA/h. Subclinical and aware-dyscognitive seizures exhibited the lowest SW-Recall and highest FA/h. MultiROCKET and XGBoost performed best overall, though no single model was optimal for all seizure types. Longer segments (60 s) generally reduced FA/h. Feature set effectiveness varied, with multi-biosignal sets improving performance across seizure types.

Conclusions

Wrist-worn wearables combined with ML can extend seizure detection beyond GTC seizures, though performance remains limited for non-motor types. Optimizing model selection, feature sets, and segment lengths, and minimizing false alarms, are key to clinical utility and real-world adoption.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Wearable Electronic DevicesSeizuresMachine LearningHumansMaleWristElectroencephalographyFemaleAdultMiddle AgedAccelerometry

Stichwörter der Autor:innen

epilepsymachine learningseizure monitoringwearable devices

Autor:innen und Institutionen

  • Robert D. and Patricia E. Kern Center for the Science of Health Care Delivery, Mayo Clinic, Rochester, MN 55905, USA.
  • Brain Neurology and Engineering Laboratory, Department of Neurology, Mayo Clinic, Rochester, MN 55905, USA.

ORCID-Profile: Louis Faust, Benjamin H Brinkmann

Förderung

  • NINDS NIH HHS
  • NCATS NIH HHS
  • the Robert D. and Patricia E. Kern Center for the Science of Health Care Delivery, the Mayo Neurology AI Program, the Mayo Center for Clinical and Translationa…
  • the Epilepsy Foundation of America’s My Seizure Gauge grant
  • the NIH

Zitieren

APA

Faust, L., Cui, J., Knepper, C., Nasseri, M., Worrell, G., & Brinkmann, B. H. (2025). Detecting Diverse Seizure Types with Wrist-Worn Wearable Devices: A Comparison of Machine Learning Approaches. Sensors, 25(17), 5562. https://doi.org/10.3390/s25175562

Vancouver

Faust L, Cui J, Knepper C, Nasseri M, Worrell G, Brinkmann BH. Detecting Diverse Seizure Types with Wrist-Worn Wearable Devices: A Comparison of Machine Learning Approaches. Sensors. 2025;25(17):5562. doi:10.3390/s25175562

BibTeX
@article{Faust2025,
  author = {Faust, Louis and Cui, Jie and Knepper, Camille and Nasseri, Mona and Worrell, Gregory and Brinkmann, Benjamin H},
  title = {Detecting Diverse Seizure Types with Wrist-Worn Wearable Devices: A Comparison of Machine Learning Approaches},
  journal = {Sensors},
  year = {2025},
  volume = {25},
  number = {17},
  pages = {5562},
  doi = {10.3390/s25175562}
}
Kostenlos starten →