• Skip to primary navigation
  • Skip to main content
  • Skip to primary sidebar
  • Skip to footer

SAGES

Reimagining surgical care for a healthier world

  • Home
    • COVID-19 Annoucements
    • Search
    • SAGES Home
    • SAGES Foundation Home
  • About
    • Who is SAGES?
    • SAGES Mission Statement
    • Advocacy
    • Strategic Plan, 2020-2023
    • Committees
      • Request to Join a SAGES Committee
      • SAGES Board of Governors
      • Officers and Representatives of the Society
      • Committee Chairs and Co-Chairs
      • Full Committee Rosters
      • SAGES Past Presidents
    • Donate to the SAGES Foundation
    • Awards
      • George Berci Award
      • Pioneer in Surgical Endoscopy
      • Excellence In Clinical Care
      • International Ambassador
      • IRCAD Visiting Fellowship
      • Social Justice and Health Equity
      • Excellence in Community Surgery
      • Distinguished Service
      • Early Career Researcher
      • Researcher in Training
      • Jeff Ponsky Master Educator
      • Excellence in Medical Leadership
      • Barbara Berci Memorial Award
      • Brandeis Scholarship
      • Advocacy Summit
      • RAFT Annual Meeting Abstract Contest and Awards
  • Meetings
    • NBT Innovation Weekend
    • SAGES Annual Meeting
      • 2023 Scientific Session Call For Abstracts
      • 2023 Emerging Technology Call For Abstracts
    • CME Claim Form
    • Industry
      • Advertising Opportunities
      • Exhibit Opportunities
      • Sponsorship Opportunities
    • Future Meetings
    • Past Meetings
      • SAGES 2022
      • SAGES 2021
    • Related Meetings Calendar
  • Join SAGES!
    • Membership Benefits
    • Membership Applications
      • Active Membership
      • Affiliate Membership
      • Associate Active Membership
      • Candidate Membership
      • International Membership
      • Medical Student Membership
    • Member News
      • Member Spotlight
      • Give the Gift of SAGES Membership
  • Patients
    • Healthy Sooner – Patient Information for Minimally Invasive Surgery
    • Patient Information Brochures
    • Choosing Wisely – An Initiative of the ABIM Foundation
    • All in the Recovery: Colorectal Cancer Alliance
    • Find a SAGES Member
  • Publications
    • SAGES Stories Podcast
    • SAGES Clinical / Practice / Training Guidelines, Statements, and Standards of Practice
    • Patient Information Brochures
    • TAVAC – Technology and Value Assessments
    • Surgical Endoscopy and Other Journal Information
    • SAGES Manuals
    • SCOPE – The SAGES Newsletter
    • COVID-19 Annoucements
    • Troubleshooting Guides
  • Education
    • OpiVoid.org
    • SAGES.TV Video Library
    • Safe Cholecystectomy Program
      • Safe Cholecystectomy Didactic Modules
    • Masters Program
      • SAGES Facebook Program Collaboratives
      • Acute Care Surgery
      • Bariatric
      • Biliary
      • Colorectal
      • Flexible Endoscopy (upper or lower)
      • Foregut
      • Hernia
      • Robotics
    • Educational Opportunities
    • HPB/Solid Organ Program
    • Courses for Residents
      • Advanced Courses
      • Basic Courses
    • Video Based Assessments (VBA)
    • Robotics Fellows Course
    • MIS Fellows Course
    • Facebook Livestreams
    • Free Webinars For Residents
    • SMART Enhanced Recovery Program
    • SAGES OR SAFETY Video
    • SAGES at Cine-Med
      • SAGES Top 21 MIS Procedures
      • SAGES Pearls
      • SAGES Flexible Endoscopy 101
      • SAGES Tips & Tricks of the Top 21
  • Opportunities
    • NEW-Area of Concentrated Training Seal (ACT)-Advanced Flexible Endoscopy-Coming Soon!
    • SAGES Fellowship Certification for Advanced GI MIS and Comprehensive Flexible Endoscopy
    • Multi-Society Foregut Fellowship Certification
    • SAGES Research Opportunities
    • Fundamentals of Laparoscopic Surgery
    • Fundamentals of Endoscopic Surgery
    • Fundamental Use of Surgical Energy
    • Job Board
    • SAGES Go Global: Global Affairs and Humanitarian Efforts
  • Search
    • Search All SAGES Content
    • Search SAGES Guidelines
    • Search the Video Library
    • Search the Image Library
    • Search the Abstracts Archive
  • Store
    • “Unofficial” Logo Products
  • Log In

Machine learning and coresets for automated, real-time video segmentation of laparoscopic surgery

Mikhail Volkov, MSc1, Daniel A Hashimoto, MD, MS2, Guy Rosman, PhD1, Ozanan Meireles, MD2, Daniela Rus1. 1MIT, 2Massachusetts General Hospital

Background and Objective: While recording laparoscopic surgical procedures is quick and easy, reviewing video remains a time-consuming process. Context-aware video segmentation must be done manually by browsing through the video to identify different steps of an operation. As an alternative, by using state-of-the-art machine learning methods trained on surgical video annotations, we demonstrate the use of coresets to automatically segment, summarize, and classify a surgical video according to its constituent steps.

Description: Coresets are compact data reduction constructs that can be used to efficiently approximate inefficient algorithms. Coreset algorithms are able to take a video stream and compute temporal segmentation of the video based on detected features. The end result is a small, hierarchical subset of video frames that summarizes the semantic content of the video. We present a system that computes such a summary automatically, online, and in real-time. This system was trained and subsequently tested on videos of laparoscopic vertical sleeve gastrectomy (LSG).

Preliminary Results: We conducted experiments on 4 surgical videos of LSG procedure, which amounted to approximately 3 hours of video footage. A set of 7 steps were thus identified for this procedure as an initial starting point by expert laparoscopic surgeons. Since not all the videos contained all the steps, we restricted initial experiments to just the major steps in this procedure, which may combine several separate steps

The expert surgeons provided qualitative semantic descriptions corresponding to their annotations which constitute the ground truth segmentation of each video sequence. By combining and augmenting standard computer vision algorithms, a new representation space was constructed based on these qualitative semantic descriptions which yields features that are highly discriminative of their respective steps.

A cascade of support vector machined (SVM) was then constructed and trained to classify each step of the surgery. We completed 3 experiments, each time leaving out an entire video as the test set, and splitting the remaining combined video frames into 80/20 training/validation set.

The system was then run live on each test video, classifying incoming frames and producing a vector of binary SVM outputs for each frame. Finally, a Hidden Markov Model (HMM) is trained and used to produce the most likely underlying states given the incoming stream of observations. 

Preliminary results show that prediction accuracy is on par with similar state of the art approaches (85%+).

Future Directions: Video-based coaching and debriefing of laparoscopic surgical procedures has been demonstrated to contribute to enhanced surgical performance; however, modern pressures on training and productivity preclude spending hours viewing and editing surgical video for the purpose of routine video-based coaching. Automated segmentation of surgical videos would allow important and interesting parts of large video sequences to be efficiently indexed for use in training and during time-critical expert consultations for targeted feedback and coaching. A summarization of the surgical video can be archived for teaching and training purposes and to help identify errors or examples of good technique that may have occurred during the procedure.

287

Share this:

  • Twitter
  • Facebook
  • LinkedIn
  • Pinterest
  • WhatsApp
  • Reddit

Related

« Return to SAGES 2016 abstract archive

Our Mission

Innovate, educate and collaborate to improve patient care.

Recently, on SAGES…

Critical View of Safety (CVS) Challenge QR Code

The SAGES Critical View of Safety Challenge – Donate Your Lap Chole Videos!

The Society of American Gastrointestinal and Endoscopic Surgeons is hosting the first Artificial Intelligence Data Challenge conducted by surgeons. The aim of this challenge is to generate a large and diverse dataset of laparoscopic cholecystectomy videos, annotated with respect to the subcomponents of the Critical View of Safety (CVS). Computer scientists from all over the […]

Respuesta de SAGES al Estudio NordICC sobre el beneficio de las colonoscopias de detección

SAGES desea aclarar los resultados del estudio NordICC y colocarlos en contexto de los esfuerzos de varias agencias nacionales para reducir el riesgo de cáncer colorrectal – la segunda causa de muerte por cáncer más frecuente en los Estados Unidos-, mediante la promoción de la detección y tratamiento oportuno de las lesiones.

SAGES Response to NordICC Study Regarding Benefit of Screening Colonoscopies

The NordICC Study recently published in The New England Journal of Medicine and widely reported on by media outlets has raised questions regarding the benefit of screening colonoscopy in lowering the risk of colorectal cancer and cancer-related deaths among otherwise healthy and symptom-free men and women aged 55 to 64. Provocative headlines and commentaries have […]

Contact SAGES

Society of American Gastrointestinal and Endoscopic Surgeons
11300 W. Olympic Blvd Suite 600
Los Angeles, CA 90064 USA
webmaster@sages.org
Tel: (310) 437-0544

Find Us Around the Web!

  • Facebook
  • Twitter
  • YouTube

Important Links

SAGES 2023 Meeting Information

Healthy Sooner: Patient Information

SAGES Guidelines, Statements, & Standards of Practice

SAGES Manuals

 

  • taTME Study Info
  • Foundation
  • SAGES.TV
  • MyCME
  • Educational Activities

Copyright © 2023 Society of American Gastrointestinal and Endoscopic Surgeons