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

Continuous sPatial-temporal deformable image registration and 4D frame interpolation.

Medical physics·2025
Same author

Cervical stiffness in mid-pregnancy: a predictive tool using a novel aspiration device for spontaneous preterm birth risk assessment.

American journal of obstetrics & gynecology MFM·2025
Same author

Gaussian primitives for deformable image registration.

Physics and imaging in radiation oncology·2025
Same author

Author Correction: Imaging and AI based chromatin biomarkers for diagnosis and therapy evaluation from liquid biopsies.

NPJ precision oncology·2025
Same author

2D echocardiography video to 3D heart shape reconstruction for clinical application.

Medical image analysis·2024
Same author

A unified generation-registration framework for improved MR-based CT synthesis in proton therapy.

Medical physics·2024

Related Experiment Video

Updated: Oct 19, 2025

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

903

Self-supervised representation learning for surgical activity recognition.

Daniel Paysan1, Luis Haug2, Michael Bajka3

  • 1Department of Computer Science, ETH Zurich, Zurich, Switzerland. paysand@ethz.ch.

International Journal of Computer Assisted Radiology and Surgery
|September 20, 2021
PubMed
Summary
This summary is machine-generated.

This study introduces a data-efficient method for surgical activity recognition in virtual reality simulators. Self-supervised learning extracts key features from surgical trajectories, improving performance in educational settings.

Keywords:
Deep LearningProbabilistic modelingRepresentation LearningSelf-supervised LearningSurgical Activity RecognitionUnsupervised Learning

More Related Videos

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

3.0K
Technical Approach for Infrared Tracking for Soft Tissue Navigation with a Holographic Head-Mounted Display and Preclinical Validation
10:25

Technical Approach for Infrared Tracking for Soft Tissue Navigation with a Holographic Head-Mounted Display and Preclinical Validation

Published on: September 2, 2025

168

Related Experiment Videos

Last Updated: Oct 19, 2025

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

903
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

3.0K
Technical Approach for Infrared Tracking for Soft Tissue Navigation with a Holographic Head-Mounted Display and Preclinical Validation
10:25

Technical Approach for Infrared Tracking for Soft Tissue Navigation with a Holographic Head-Mounted Display and Preclinical Validation

Published on: September 2, 2025

168

Area of Science:

  • Medical Education Technology
  • Computer Science
  • Artificial Intelligence

Background:

  • Virtual reality (VR) simulators are crucial for surgical education.
  • Automatic recognition and assessment of surgical activities are needed for effective VR training.
  • Current methods require expensive, expert-annotated data, limiting scalability.

Purpose of the Study:

  • To develop a data-efficient method for surgical activity recognition in VR simulators.
  • To leverage self-supervised learning for feature extraction from surgical trajectories.
  • To improve the assessment of surgical skills in simulated environments.

Main Methods:

  • Utilized self-supervised training of deep encoder-decoder architectures.
  • Learned representations from surgical video data to capture trajectory events.
  • Employed unsupervised pipelines for surgical activity recognition using extracted features.

Main Results:

  • Features from deep representation learning enhanced hidden semi-Markov models for activity recognition.
  • Demonstrated improved performance in a simulated myomectomy scenario.
  • Validated the benefit of predicting surgery progress for feature extraction.

Conclusions:

  • This work is a significant step towards efficient feature utilization in surgical activity recognition.
  • Enables effective use of deep representation learning with limited annotated data.
  • Addresses challenges of incomplete and expensive annotations in surgical training data.