Development and Validation of Accelerometer-Based Machine Learning Models for Classifying Walking, Running, and Jumping Activities
Sensors · 30.04.2026 · 26 (9) · 2810
DOI: 10.3390/s26092810 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sensors, online erschienen am 30.04.2026. Verfasst von Lucas Veras und 5 weiteren Autor:innen. Beteiligte Einrichtung: Research Center in Physical Activity, Health and Leisure (CIAFEL), Faculty of Sport, University of Porto, 4200-450 Porto, Portugal. Hauptthemen laut MEDLINE: Accelerometry, Running, Machine Learning, Walking. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 42 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Fundação para a Ciência e Tecnologia, Laboratory for Integrative and Translational Research in Population Health.
Abstract (Originalsprache)
Quantifying mechanical loading during daily physical activities is essential for designing and evaluating bone health interventions. Accelerometers are a promising tool for estimating these loads under free-living conditions, yet existing prediction models depend on prior knowledge of the activity being performed. This study developed and validated machine learning models to automatically distinguish between walking, running, and jumping using accelerometer data. Forty-eight healthy adults completed a protocol of walking, running, and jumping tasks while wearing ActiGraph GT9X Link accelerometers at the ankle, lower back, and hip. Three algorithms (Random Forest, Support Vector Machine, and K-Nearest Neighbors) were trained and evaluated through multiple performance metrics. All models achieved excellent classification accuracy across sensor placements, with percent agreement between 93.8% and 97.7%, receiver operating characteristic area under the curve values consistently above 0.97, and Kappa coefficients exceeding 0.89. These results demonstrate that accelerometer-based activity classification can reliably differentiate walking, running, and jumping, establishing a practical framework for applying activity-specific mechanical loading prediction equations under free-living conditions.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Research Center in Physical Activity, Health and Leisure (CIAFEL), Faculty of Sport, University of Porto, 4200-450 Porto, Portugal.
- Laboratory for Integrative and Translational Research in Population Health (ITR), University of Porto, 4200-450 Porto, Portugal.
- Institute of Molecular Sports and Rehabilitation Medicine, Paracelsus Medical Private University, 5020 Salzburg, Austria.
- Centro Regional de Estudios Avanzados en Estilos de Vida Activos y Saludables (CREA-EVAS), Universidad Adventista de Chile, Chillán 3780000, Chile.
ORCID-Profile: Lucas Veras, Giorjines Boppre, Ana Resende-Coelho, José Oliveira, Hélder Fonseca
Förderung
- Fundação para a Ciência e Tecnologia
- Laboratory for Integrative and Translational Research in Population Health
Zitieren
APA
Veras, L., Diniz-Sousa, F., Boppre, G., Resende-Coelho, A., Oliveira, J., & Fonseca, H. (2026). Development and Validation of Accelerometer-Based Machine Learning Models for Classifying Walking, Running, and Jumping Activities. Sensors, 26(9), 2810. https://doi.org/10.3390/s26092810
Vancouver
Veras L, Diniz-Sousa F, Boppre G, Resende-Coelho A, Oliveira J, Fonseca H. Development and Validation of Accelerometer-Based Machine Learning Models for Classifying Walking, Running, and Jumping Activities. Sensors. 2026;26(9):2810. doi:10.3390/s26092810
BibTeX
@article{Veras2026,
author = {Veras, Lucas and Diniz-Sousa, Florêncio and Boppre, Giorjines and Resende-Coelho, Ana and Oliveira, José and Fonseca, Hélder},
title = {Development and Validation of Accelerometer-Based Machine Learning Models for Classifying Walking, Running, and Jumping Activities},
journal = {Sensors},
year = {2026},
volume = {26},
number = {9},
pages = {2810},
doi = {10.3390/s26092810}
}
EUVIMED