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 Experiment Video

Updated: May 27, 2026

Automatic Surgery in Transcatheter Aortic Valve Replacement Using Augmented Reality
07:46

Automatic Surgery in Transcatheter Aortic Valve Replacement Using Augmented Reality

Published on: August 9, 2024

Cross-Center Surgical Step Recognition in Standardized Training Tasks: Dataset, Baselines, and Transfer Analysis.

Georg Wolf, Alexandra Eberenz, Georg Mannel

    IEEE Transactions on Bio-Medical Engineering
    |May 25, 2026
    PubMed
    Summary
    This summary is machine-generated.

    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

    [Preventing Complications in Hepatopancreatobiliary Surgery].

    Zentralblatt fur Chirurgie·2026
    Same author

    Feasibility of Vibroacoustic Sensing for Detection of Peritoneal Entry During Laparoscopic Access: A Pilot Study in a Human Body Donor.

    Diagnostics (Basel, Switzerland)·2026
    Same author

    Effect of multimodal therapy in the treatment of adenocarcinoma of the esophagus and gastroesophageal junction : a population-based analysis from the German Clinical Cancer Registry Group.

    Langenbeck's archives of surgery·2026
    Same author

    Impact of single freeze-thaw cycles on human serum proteins: Implications for mass spectrometry biomarker validation.

    iScience·2026
    Same author

    Activity and Physiological Stress Within 90 Days After Minimally Invasive and Open Pancreatoduodenectomy: A Predefined Analysis of the DIPLOMA-2 Randomized Clinical Trial.

    JAMA surgery·2026
    Same author

    Robotic Pancreatoduodenectomy Carries Greater Risk of Morbidity after Conversion than Laparoscopy: A Propensity Score-Matched Analysis in the E-MIPS Registry.

    Annals of surgery·2026

    Automated surgical step recognition in box trainers is crucial for feedback. Self-supervised learning, particularly temporal-order tasks, significantly improves cross-center performance with minimal data.

    Area of Science:

    • Robotics and Automation in Surgery
    • Medical Education Technology
    • Computer Vision for Healthcare

    Background:

    • Automated assessment of surgical skills is vital for enhancing training and patient safety.
    • Step-specific feedback requires reliable surgical step recognition, which is underexplored in box-trainer exercises.
    • Methods must be robust to variations in platforms, instruments, and cameras for practical deployment.

    Purpose of the Study:

    • To formalize surgical steps for box-trainer exercises and establish a benchmark for cross-center step recognition.
    • To evaluate self-supervised pretraining schemes and finetuning strategies for robust surgical step recognition.
    • To release the first Multi-center Surgical Training dataset (MiST-STEP) with step annotations.

    Main Methods:

    Related Experiment Videos

    Last Updated: May 27, 2026

    Automatic Surgery in Transcatheter Aortic Valve Replacement Using Augmented Reality
    07:46

    Automatic Surgery in Transcatheter Aortic Valve Replacement Using Augmented Reality

    Published on: August 9, 2024

  • Formalized surgical steps for three box-trainer exercises.
  • Developed and benchmarked a two-stage pipeline for cross-center step recognition.
  • Investigated two self-supervised pretraining methods (MoCo v2, temporal-order task) and three finetuning strategies.

    • Main Results:

    • Single-center models exhibited significant performance degradation (up to 48 pp macro-F1) on unseen centers.
    • Temporal-order pretraining outperformed MoCo v2, reducing the performance gap by two-thirds.
    • Finetuning with as few as two labeled videos per target center largely closed the remaining performance gap.

    Conclusions:

    • This study presents the first cross-center benchmark for surgical step recognition in training.
    • Self-supervision benefits are task-dependent; temporal-order tasks show promise for surgical training.
    • Center-robust step recognition is key for automated feedback systems to standardize training and improve patient outcomes.