EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Self-Supervised Transfer Learning for IMU-Based Upper-Limb Action Detection and Motion Quality Analysis in an Immersive VR Functional Task

Zhao Liu, Daniele Soria, Chee Siang Ang, Sukhi Shergill

Journal of Sensor and Actuator Networks · 12.06.2026 · 15 (3) · 46

DOI: 10.3390/jsan15030046 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Sensor and Actuator Networks, online erschienen am 12.06.2026. Verfasst von Zhao Liu und 3 weiteren Autor:innen. Die Arbeit nennt 54 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: University of Kent.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Wearable inertial sensing has considerable potential for process-level analysis of upper-limb function, but further evidence is needed to understand how it can be applied within ecologically structured immersive virtual reality (VR) tasks. Most VR-based functional assessments rely primarily on outcome-level indicators, such as task completion time, success rate, or error count, which may not fully capture how a task is executed. This exploratory study investigated whether wearable IMU signals collected during an immersive VR sushi-making task could support binary detection of a core upper-limb manipulation phase and provide additional information about task execution beyond global performance outcomes. A total of 45 participants contributed usable motion recordings for this study, with five Xsens DOT sensors placed on the hands, forearms, and waist. Three signal modalities were analysed, including acceleration (ACC), gyroscope angular velocity (GYR), and Euler angles. The downstream recognition problem was formulated as a binary classification task (Placing vs. Non-Placing), and a self-supervised learning (SSL) pretrain–fine-tune strategy was evaluated against conventional machine learning and from-scratch deep learning baselines using five subject-wise validation splits. The strongest overall performance was achieved with hand-mounted accelerometer signals, with LeftHand–ACC achieving a Macro-F1 of 0.712±0.128 and RightHand–ACC achieving 0.679±0.118. Under both hand-ACC settings, SSL fine-tuning showed higher mean Macro-F1 than the Balanced Random Forest baseline and the same deep architecture trained from scratch. Recognition performance varied substantially across sensor locations, signal modalities, and task segments, with distal upper-limb sensors generally outperforming waist-based configurations. Cross-age analyses further showed that within-cohort and cross-cohort performance did not fully align, indicating sensitivity to age-related distribution shift. Beyond classification, Log Dimensionless Jerk (LDLJ) derived from the Placing action showed a significant positive association with Cognitron motor control time cost (r=0.636, p

Förderung

  • University of Kent

Zitieren

APA

Liu, Z., Soria, D., Ang, C. S., & Shergill, S. (2026). Self-Supervised Transfer Learning for IMU-Based Upper-Limb Action Detection and Motion Quality Analysis in an Immersive VR Functional Task. Journal of Sensor and Actuator Networks, 15(3), 46. https://doi.org/10.3390/jsan15030046

Vancouver

Liu Z, Soria D, Ang CS, Shergill S. Self-Supervised Transfer Learning for IMU-Based Upper-Limb Action Detection and Motion Quality Analysis in an Immersive VR Functional Task. Journal of Sensor and Actuator Networks. 2026;15(3):46. doi:10.3390/jsan15030046

BibTeX
@article{Liu2026,
  author = {Liu, Zhao and Soria, Daniele and Ang, Chee Siang and Shergill, Sukhi},
  title = {Self-Supervised Transfer Learning for IMU-Based Upper-Limb Action Detection and Motion Quality Analysis in an Immersive VR Functional Task},
  journal = {Journal of Sensor and Actuator Networks},
  year = {2026},
  volume = {15},
  number = {3},
  pages = {46},
  doi = {10.3390/jsan15030046}
}
Kostenlos starten →