EUVIMEDEUVIMED entdecken
Crossref

STANet: A Surgical Gesture Recognition Method Based on Spatiotemporal Fusion

Boqiang Jia, Wenjie Wang iD, Xin Tian, Xiaohua Wang

Annals of the New York Academy of Sciences · 2025-11 · 1553 (1) · 502-516

DOI: 10.1111/nyas.70053 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Annals of the New York Academy of Sciences, online erschienen am 24.09.2025. Verfasst von Boqiang Jia und 3 weiteren Autor:innen. Beteiligte Einrichtung: School of Electronics and Information, Xi'an Polytechnic University, Xi'an, Shaanxi, China. Hauptthemen laut MEDLINE: Gestures, Robotic Surgical Procedures, Pattern Recognition, Automated. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 58 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Innovation Ability Promotion Project of Shandong Province, Key Project of Natural Science Basic Research Program of Shaanxi.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In robotic surgery, surgical gesture recognition has great importance in surgical quality evaluation and intelligent recognition assistance. Currently, deep learning models, such as recurrent neural networks and temporal convolutional networks, are mainly used to model action sequences and capture the temporal dependencies between them. However, some of these methods ignore the fusion of spatial and temporal features, and hence cannot effectively capture long‐term relationships and efficiently model action sequences. To overcome these limitations, we propose a spatiotemporal adaptive network (STANet) to fuse spatiotemporal features. Specifically, we designed a temporal module and a spatial module to extract respective features. Subsequently, these features were fused and further refined through temporal modeling using a temporal adaptive convolution strategy. This approach integrates both long‐term and short‐term characteristics of surgical gesture sequences. The organic combination of temporal and spatial modules was inserted into the backbone network to form the STANet, which efficiently modeled the action sequences. Our approach has been validated on the publicly available surgical gesture datasets JIGSAWS and RARP‐45, achieving very good results. Compared to other reported benchmark models, our model demonstrates exceptional performance. It can be used in surgical robots, visual feedback systems, and computer‐assisted surgery.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

GesturesRobotic Surgical ProceduresPattern Recognition, AutomatedHumansNeural Networks, ComputerDeep LearningAlgorithms

Stichwörter der Autor:innen

deep learningrobotic surgerysurgery informationsurgical gesture classificationtime series analysis

Autor:innen und Institutionen

  • School of Electronics and Information, Xi'an Polytechnic University, Xi'an, Shaanxi, China.
  • Xi'an Polytechnic University Branch of Shaanxi Artificial Intelligence Joint Laboratory, Xi'an Polytechnic University, Xi'an, Shaanxi, China.

ORCID-Profile: Wenjie Wang

Förderung

  • National Natural Science Foundation of China
  • Innovation Ability Promotion Project of Shandong Province
  • Key Project of Natural Science Basic Research Program of Shaanxi
  • Xi'an Polytechnic University Graduate Innovation Fund Project

Zitieren

APA

Jia, B., Wang, W., Tian, X., & Wang, X. (2025). STANet: A Surgical Gesture Recognition Method Based on Spatiotemporal Fusion. Annals of the New York Academy of Sciences, 1553(1), 502-516. https://doi.org/10.1111/nyas.70053

Vancouver

Jia B, Wang W, Tian X, Wang X. STANet: A Surgical Gesture Recognition Method Based on Spatiotemporal Fusion. Annals of the New York Academy of Sciences. 2025;1553(1):502-516. doi:10.1111/nyas.70053

BibTeX
@article{Jia2025,
  author = {Jia, Boqiang and Wang, Wenjie and Tian, Xin and Wang, Xiaohua},
  title = {STANet: A Surgical Gesture Recognition Method Based on Spatiotemporal Fusion},
  journal = {Annals of the New York Academy of Sciences},
  year = {2025},
  volume = {1553},
  number = {1},
  pages = {502--516},
  doi = {10.1111/nyas.70053}
}
Kostenlos starten →