EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

ParCuR—A Novel AI-Enabled Gait Cueing Wearable for Patients with Parkinson’s Disease

Telmo Lopes iD, Manuel Reis Carneiro iD, Ana Morgadinho iD, Diogo Reis Carneiro iD, Mahmoud Tavakoli iD

Sensors · 2025

DOI: 10.3390/s25227077 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, 2025. Verfasst von Telmo Lopes und 4 weiteren Autor:innen. Beteiligte Einrichtung: Institute of Systems and Robotics, Department of Electrical and Computer Engineering, University of Coimbra, 3030-290 Coimbra, Portugal. Hauptthemen laut MEDLINE: Parkinson Disease, Wearable Electronic Devices, Gait, Artificial Intelligence. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Fundação para a Ciência e Tecnologia, European Research Council, Fundação para a Ciência e a Tecnologia.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Freezing of gait (FoG) is a common motor symptom in advanced Parkinson’s disease, leading to falls, disability, and reduced quality of life. Although cueing systems using visual or auditory stimuli can help patients resume walking, existing solutions are often expensive, uncomfortable, and conspicuous. ParCuR (Parkinson Cueing and Rehabilitation) is a compact, ankle-worn wearable integrating an inertial sensor, haptic stimulator, and AI-based software. It was developed to detect FoG episodes in real time and provides automatic sensory cues to assist patients with Parkinson’s Disease (PwP). A classifier was trained for FoG detection using the DAPHNet dataset, comparing patient-specific and patient-independent models. While a small-scale trial with PwP assessed usability and reliability. ParCuR is watch-sized (35 × 41 mm), discreet, and comfortable for daily use. The online detection algorithm triggers stimulation within 0.7 s of episode onset and achieves 94.9% sensitivity and 91.3% specificity using only 14 frequency-based features. Preliminary trials confirmed device feasibility and guided design refinements. This low-cost, wearable solution supports personalized, real-time FoG detection and responsive cueing, improving patient mobility while minimizing discomfort and continuous stimulation habituation.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Parkinson DiseaseWearable Electronic DevicesGaitArtificial IntelligenceHumansMaleCuesAgedFemaleMiddle AgedAlgorithmsGait Disorders, Neurologic

Stichwörter der Autor:innen

Parkinson’s diseaseassistive cueingfreezing of gaitmachine learningsomatosensory stimulationwearable electronics

Autor:innen und Institutionen

  • Institute of Systems and Robotics, Department of Electrical and Computer Engineering, University of Coimbra, 3030-290 Coimbra, Portugal.
  • Department of Health Sciences and Technology, ETH Zurich, 8092 Zurich, Switzerland.
  • Neurology Department, Centro Hospitalar e Universitário de Coimbra, 3004-561 Coimbra, Portugal.
  • Faculty of Medicine, University of Coimbra, 3000-548 Coimbra, Portugal.

ORCID-Profile: Telmo Lopes, Manuel Reis Carneiro, Ana Morgadinho, Diogo Reis Carneiro, Mahmoud Tavakoli

Förderung

  • Fundação para a Ciência e Tecnologia
  • European Research Council
  • Fundação para a Ciência e a Tecnologia
  • CMU-Portugal project WoW
  • European Research Council, ERC project Liquid3D

Zitieren

APA

Lopes, T., Reis Carneiro, M., Morgadinho, A., Reis Carneiro, D., & Tavakoli, M. (2025). ParCuR—A Novel AI-Enabled Gait Cueing Wearable for Patients with Parkinson’s Disease. Sensors. https://doi.org/10.3390/s25227077

Vancouver

Lopes T, Reis Carneiro M, Morgadinho A, Reis Carneiro D, Tavakoli M. ParCuR—A Novel AI-Enabled Gait Cueing Wearable for Patients with Parkinson’s Disease. Sensors. 2025. doi:10.3390/s25227077

BibTeX
@article{Lopes2025,
  author = {Lopes, Telmo and Reis Carneiro, Manuel and Morgadinho, Ana and Reis Carneiro, Diogo and Tavakoli, Mahmoud},
  title = {ParCuR—A Novel AI-Enabled Gait Cueing Wearable for Patients with Parkinson’s Disease},
  journal = {Sensors},
  year = {2025},
  doi = {10.3390/s25227077}
}
Kostenlos starten →