Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Accurate and task-agnostic modeling of enzymatic reactions through multimodal relational learning.

Acta pharmaceutica Sinica. B·2026
Same author

Capturing omega sign in the clinical assessment of depression by deep learning.

Translational psychiatry·2026
Same author

EndoLRMGS: Combining Large Reconstruction Modelling and Gaussian Splatting for Complete Endoscopic Scene Reconstruction.

IEEE transactions on medical imaging·2026
Same author

Robust quantification of ICG fluorescence perfusion in neonatal bowel surgery via deep point tracking.

International journal of computer assisted radiology and surgery·2026
Same author

An AI-Based OCT System to Detect Diabetic Macular Edema: A Prospective Validation and Noninferiority Randomized Clinical Trial.

JAMA·2026
Same author

Blob representation of robotic surgical scenes for position-aware video generation.

International journal of computer assisted radiology and surgery·2026

Related Experiment Video

Updated: Aug 28, 2025

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

485

Trans-SVNet: hybrid embedding aggregation Transformer for surgical workflow analysis.

Yueming Jin1, Yonghao Long2, Xiaojie Gao2

  • 1Wellcome/EPSRC Centre for Interventional and Surgical Sciences (WEISS), Department of Computer Science, University College, London, UK.

International Journal of Computer Assisted Radiology and Surgery
|September 21, 2022
PubMed
Summary

This study introduces Trans-SVNet, a novel Transformer-based model for surgical workflow analysis. Trans-SVNet enhances recognition and anticipation by effectively integrating spatial and temporal features, outperforming existing methods in real-time applications.

Keywords:
Spatial–temporal feature modelingSurgical visionTransformerWorkflow anticipationWorkflow recognition

More Related Videos

Author Spotlight: Revolutionizing Remote Surgery with Augmented Reality and Robotics for Enhanced Precision and Accessibility
07:46

Author Spotlight: Revolutionizing Remote Surgery with Augmented Reality and Robotics for Enhanced Precision and Accessibility

Published on: August 9, 2024

832
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.9K

Related Experiment Videos

Last Updated: Aug 28, 2025

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

485
Author Spotlight: Revolutionizing Remote Surgery with Augmented Reality and Robotics for Enhanced Precision and Accessibility
07:46

Author Spotlight: Revolutionizing Remote Surgery with Augmented Reality and Robotics for Enhanced Precision and Accessibility

Published on: August 9, 2024

832
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.9K

Area of Science:

  • Computer-assisted intervention systems
  • Surgical workflow analysis
  • Machine learning for healthcare

Background:

  • Current surgical workflow analysis methods often lose critical intermediate features due to conventional temporal modeling.
  • Existing spatial-temporal feature encoding limits the comprehensive understanding of surgical procedures.
  • There is a need for advanced models to improve cognitive assistance in real-time surgical interventions.

Purpose of the Study:

  • To develop a novel method for surgical workflow analysis that effectively preserves critical information for recognition and anticipation.
  • To rethink feature encoding strategies by exploring the complementary effects of spatial and temporal representations.
  • To improve the accuracy and efficiency of real-time computer-assisted intervention systems.

Main Methods:

  • Introduction of a hybrid embedding aggregation Transformer, named Trans-SVNet.
  • Utilizing spatial embeddings to query temporal embedding sequences for integrated feature interaction.
  • Joint optimization using loss objectives from both workflow recognition and anticipation tasks.

Main Results:

  • Trans-SVNet consistently outperforms state-of-the-art methods on three large surgical video datasets for workflow recognition.
  • Jointly learning recognition and anticipation tasks significantly improves recognition performance.
  • The model demonstrates promising effectiveness and efficiency, achieving real-time inference speed.

Conclusions:

  • The hybrid embedding integration in Trans-SVNet effectively captures crucial cues from complementary spatial-temporal data.
  • Multi-task learning, incorporating anticipation, enhances the knowledge transfer for the recognition task.
  • Trans-SVNet shows significant potential for real-world application in operating rooms due to its effectiveness and efficiency.