EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Dual-Stream deep learning for multimodal feature fusion and classification of balance control in elite freestyle aerial skiers

Xinze Cui, Jie Gao, Pengquan Zhang, Jingyi Yan, Yongxia Chen, Guocai Xu, Yuqi Cheng, Xin Wang iD, Yanming Fu

PLOS One · 28.07.2026 · 21 (7) · e0337296

DOI: 10.1371/journal.pone.0337296 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, online erschienen am 28.07.2026. Verfasst von Xinze Cui und 8 weiteren Autor:innen. Beteiligte Einrichtung: School of Physical Education, Liaoning Normal University, Dalian, China. Hauptthemen laut MEDLINE: Skiing, Postural Balance, Deep Learning. Die Arbeit nennt 55 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Balance control is a key determinant of stable landing in elite freestyle aerial skiing. Rapid and precise identification of subtle differences in athletes’ balance-stability regulation is a prerequisite for targeted, evidence-based training. Conventional balance assessment typically relies on force-platform measurements of the center of pressure (COP) trajectory and subsequent time-, frequency-, and time–frequency–domain analyses. However, these indices have limited ability to capture the complex dynamics of postural control and to discriminate fine-scale differences in balance regulation among highly trained freestyle skiing aerials athletes.To address this limitation, we developed a dual-stream deep learning model that fuses time–frequency image features with COP-based statistical descriptors to classify subtle variations in balance regulation. Twenty-five elite freestyle skiing aerials athletes were recruited and performed quiet standing under two conditions: (i) bipedal stance on a stable surface with eyes open and (ii) bipedal stance on an unstable surface with eyes open. COP trajectories were recorded and their multiscale entropy computed; K-means clustering was used to stratify participants into high-, medium-, and low-stability groups. The extracted time–frequency and statistical features were then fed into the dual-stream deep learning framework for model training and validation.The proposed model achieved approximately 95% classification accuracy in distinguishing data-driven COP-based stability strata, suggesting potential utility for the sensitive assessment of balance-regulation patterns in elite freestyle skiing aerials athletes.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

SkiingPostural BalanceDeep LearningHumansMaleAthletesAdultYoung AdultFemale

Autor:innen und Institutionen

  • School of Physical Education, Liaoning Normal University, Dalian, China.
  • College of Exercise Health, Shenyang Sport University, Shenyang, China.
  • College of Winter Sports, Shenyang Sport University, Shenyang, China.
  • School of Physical Education and Leisure, Guangdong Ocean University, Zhanjiang, China.
  • College of Exercise Science and Health, Harbin Sport University, Harbin, China.

ORCID-Profile: Xin Wang

Zitieren

APA

Cui, X., Gao, J., Zhang, P., Yan, J., Chen, Y., Xu, G., Cheng, Y., Wang, X., & Fu, Y. (2026). Dual-Stream deep learning for multimodal feature fusion and classification of balance control in elite freestyle aerial skiers. PLOS One, 21(7), e0337296. https://doi.org/10.1371/journal.pone.0337296

Vancouver

Cui X, Gao J, Zhang P, Yan J, Chen Y, Xu G, et al. Dual-Stream deep learning for multimodal feature fusion and classification of balance control in elite freestyle aerial skiers. PLOS One. 2026;21(7):e0337296. doi:10.1371/journal.pone.0337296

BibTeX
@article{Cui2026,
  author = {Cui, Xinze and Gao, Jie and Zhang, Pengquan and Yan, Jingyi and Chen, Yongxia and Xu, Guocai and Cheng, Yuqi and Wang, Xin and Fu, Yanming},
  title = {Dual-Stream deep learning for multimodal feature fusion and classification of balance control in elite freestyle aerial skiers},
  journal = {PLOS One},
  year = {2026},
  volume = {21},
  number = {7},
  pages = {e0337296},
  doi = {10.1371/journal.pone.0337296}
}
Kostenlos starten →