EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Action recognition in rehabilitation: combining 3D convolution and LSTM with spatiotemporal attention

Fan Yang, Shiyu Li, Chang Sun, Xingjiang Li, Zhangbo Xiao

Frontiers in Physiology · 02.12.2024 · 15

DOI: 10.3389/fphys.2024.1472380 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Physiology, online erschienen am 02.12.2024. Verfasst von Fan Yang und 4 weiteren Autor:innen. Beteiligte Einrichtung: The Second Affiliated Hospital, Qiqihar Medical University, Qiqihar, Heilongjiang, China. Die Arbeit wurde bisher 14-mal zitiert (Crossref) und nennt 41 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This study addresses the limitations of traditional sports rehabilitation, emphasizing the need for improved accuracy and response speed in real-time action detection and recognition in complex rehabilitation scenarios. We propose the STA-C3DL model, a deep learning framework that integrates 3D Convolutional Neural Networks (C3D), Long Short-Term Memory (LSTM) networks, and spatiotemporal attention mechanisms to capture nuanced action dynamics more precisely. Experimental results on multiple datasets, including NTU RGB + D, Smarthome Rehabilitation, UCF101, and HMDB51, show that the STA-C3DL model significantly outperforms existing methods, achieving up to 96.42% accuracy and an F1 score of 95.83% on UCF101, with robust performance across other datasets. The model demonstrates particular strength in handling real-time feedback requirements, highlighting its practical application in enhancing rehabilitation processes. This work provides a powerful, accurate tool for action recognition, advancing the application of deep learning in rehabilitation therapy and offering valuable support to therapists and researchers. Future research will focus on expanding the model’s adaptability to unconventional and extreme actions, as well as its integration into a wider range of rehabilitation settings to further support individualized patient recovery.

Schlagwörter

Stichwörter der Autor:innen

action recognitionbiofeedback systemdata analysisdeep learningmotion rehabilitationneural networks

Autor:innen und Institutionen

  • The Second Affiliated Hospital, Qiqihar Medical University, Qiqihar, Heilongjiang, China.
  • The Second Affiliated Hospital, Pingfang District People's Hospital, Harbin, China.
  • Department of Health and Care, Qiqihar Institute of Engineering, Qiqihar, China.
  • Pathology College, Qiqihar Medical University, Qiqihar, Heilongjiang, China.

Zitieren

APA

Yang, F., Li, S., Sun, C., Li, X., & Xiao, Z. (2024). Action recognition in rehabilitation: combining 3D convolution and LSTM with spatiotemporal attention. Frontiers in Physiology, 15. https://doi.org/10.3389/fphys.2024.1472380

Vancouver

Yang F, Li S, Sun C, Li X, Xiao Z. Action recognition in rehabilitation: combining 3D convolution and LSTM with spatiotemporal attention. Frontiers in Physiology. 2024;15. doi:10.3389/fphys.2024.1472380

BibTeX
@article{Yang2024,
  author = {Yang, Fan and Li, Shiyu and Sun, Chang and Li, Xingjiang and Xiao, Zhangbo},
  title = {Action recognition in rehabilitation: combining 3D convolution and LSTM with spatiotemporal attention},
  journal = {Frontiers in Physiology},
  year = {2024},
  volume = {15},
  doi = {10.3389/fphys.2024.1472380}
}
Kostenlos starten →