EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Semi-Supervised Learning for Time-Series Segmentation in Equine Gait Event Detection

Mahaut Gerard, Guillaume Dubois, Sandrine Hanne-Poujade, Camille Hebert, Henry Chateau, Neila Mezghani

IEEE Access · 2026 · 14 · 68815-68829

DOI: 10.1109/access.2026.3688918 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Access, 2026. Verfasst von Mahaut Gerard und 5 weiteren Autor:innen. Die Arbeit nennt 44 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Agence Nationale de la Recherche et de la Technologie (ANRT), France, Research Funding from the Région Nouvelle-Aquitaine, France, Institutional Funding from French Ministry of Agriculture, France.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Locomotor injuries in horses impair both welfare and performance. Traditional visual assessments of equine locomotion by veterinarians are inherently subjective. To provide objective evaluation, systems using Inertial Measurement Units (IMUs) have been developed, enabling quantitative analysis by calculating trunk symmetry indices (head, withers, pelvis) during trot. Stance temporal parameters provide valuable information for locomotion analysis, but their accurate estimation requires precise detection of stance-related events. Gait events are typically identified using signal processing techniques applied to data from limb-mounted IMUs, often placed on the cannon bones. Despite their potential, these sensors are not well tolerated by all horses and the associated signal processing methods often fail to generalize across diverse locomotion patterns. A large amount of data is gathered by veterinarians in uncontrolled environments, opening the door for deep learning methods, but the annotation process needed for supervised learning remains time-consuming and costly. The aim of this study was to segment horse stances using trunk-mounted IMU data and to introduce two novel semi-supervised time-series segmentation methods, specifically tailored for contexts with limited strongly labeled data and abundant weakly labeled data. The best performing pipeline achieved errors of $- 0.1\pm 1.6$ % of the stride duration for foot-on and $0.2\pm 2.6$ % for heel-off events of the right forelimb, and $0.2\pm 1.5$ % for foot-on and $- 0.4\pm 1.4$ % for heel-off events of the right hindlimb. Beyond improving stance segmentation in real clinical settings, this work explores two semi-supervised methods for time-series segmentation.

Förderung

  • Agence Nationale de la Recherche et de la Technologie (ANRT), France
  • Research Funding from the Région Nouvelle-Aquitaine, France
  • Institutional Funding from French Ministry of Agriculture, France

Zitieren

APA

Gerard, M., Dubois, G., Hanne-Poujade, S., Hebert, C., Chateau, H., & Mezghani, N. (2026). Semi-Supervised Learning for Time-Series Segmentation in Equine Gait Event Detection. IEEE Access, 14, 68815-68829. https://doi.org/10.1109/access.2026.3688918

Vancouver

Gerard M, Dubois G, Hanne-Poujade S, Hebert C, Chateau H, Mezghani N. Semi-Supervised Learning for Time-Series Segmentation in Equine Gait Event Detection. IEEE Access. 2026;14:68815-68829. doi:10.1109/access.2026.3688918

BibTeX
@article{Gerard2026,
  author = {Gerard, Mahaut and Dubois, Guillaume and Hanne-Poujade, Sandrine and Hebert, Camille and Chateau, Henry and Mezghani, Neila},
  title = {Semi-Supervised Learning for Time-Series Segmentation in Equine Gait Event Detection},
  journal = {IEEE Access},
  year = {2026},
  volume = {14},
  pages = {68815--68829},
  doi = {10.1109/access.2026.3688918}
}
Kostenlos starten →