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 Videos

Phase Segmentation Methods for an Automatic Surgical Workflow Analysis.

Dinh Tuan Tran1, Ryuhei Sakurai2, Hirotake Yamazoe2

  • 1Graduate School of Information Science and Engineering, Ritsumeikan University, Kusatsu, Shiga, Japan.

International Journal of Biomedical Imaging
|April 15, 2017
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

Analysis of head and chest movements that correspond to gaze directions during walking.

Experimental brain research·2019
Same journal

Deep Learning for Brain Tumour Analysis: A Systematic Review of CNN-Transformer Hybrids in Multimodal Imaging.

International journal of biomedical imaging·2026
Same journal

Brain Tumor Segmentation Using U-Net With ResNet50 Encoder for Enhanced MRI Analysis.

International journal of biomedical imaging·2026
Same journal

Generative AI-Driven CNN Framework for Enhanced Lung Cancer Detection, Prediction, and Treatment: A Novel Approach to Overcoming AI Limitations.

International journal of biomedical imaging·2026
Same journal

Enhancing the Generalizability of Deep Learning-Based Models for Lung Field Segmentation in Chest Radiographs Using Edge-Assisted Multiscale Feature Fusion.

International journal of biomedical imaging·2026
Same journal

Personalized PET Imaging in Gastric Cancer: An Umbrella Review of Meta-Analyses to Guide Radiopharmaceutical Selection and Clinical Indication.

International journal of biomedical imaging·2026
Same journal

Clinician-Centric Explainable Artificial Intelligence Framework for Medical Imaging Diagnostics: A Systematic Review.

International journal of biomedical imaging·2026
See all related articles

This study introduces robust methods for surgical phase recognition using latent Dirichlet allocation (LDA) and hidden Markov models (HMM). The approach achieved 84.4% accuracy in segmenting surgical workflows, promising for operating room applications.

Area of Science:

  • Computer Vision
  • Medical Informatics
  • Machine Learning

Background:

  • Automated surgical phase recognition is crucial for improving operating room efficiency and patient safety.
  • Current methods often lack robustness in segmenting complex surgical workflows.
  • Real-time phase identification can enable context-aware assistance and performance analysis.

Purpose of the Study:

  • To develop and validate robust methods for automatic segmentation of surgical phases within a workflow.
  • To accurately label each time point of a surgical procedure in an operating room setting.
  • To leverage optical flow motion features for enhanced workflow understanding.

Main Methods:

  • Utilized latent Dirichlet allocation (LDA) for topic modeling of optical flow motion features.

Related Experiment Videos

  • Constructed a hidden Markov model (HMM) based on LDA outputs to represent surgical workflows.
  • Employed multiple synchronized cameras to capture comprehensive working contexts, including staff, equipment, and materials.
  • Validated methods through experiments on surgical workflows with up to 12 distinct phases.

    • Main Results:

    • Achieved a maximum average accuracy of 84.4% using leave-one-out cross-validation.
    • Demonstrated robustness in segmenting surgical workflows with an average length of 12.8 minutes.
    • The combined LDA-HMM approach effectively captured motion dynamics for phase recognition.

    Conclusions:

    • The proposed LDA-HMM framework offers a promising solution for automated surgical phase segmentation.
    • High accuracy rates indicate the potential for real-world application in operating rooms.
    • This technology can pave the way for advanced surgical analytics and intelligent assistance systems.