Related Experiment Video
Updated: Jun 25, 2025
07:27
Robotics in Surgery: A Modular Robotic Platform Driven Gastric Wedge Resection
Published on: February 7, 2025
440
Laparoscopic distal gastrectomy skill evaluation from video: a new artificial intelligence-based instrument
Shiro Matsumoto1, Hiroshi Kawahira2, Kyohei Fukata3
1Department of Surgery, Division of Gastroenterological, General and Transplant Surgery, Jichi Medical University, Tochigi, Japan. s-matsumoto@jichi.ac.jp.
Scientific Reports
|May 30, 2024
Summary
Artificial intelligence (AI) object detection analyzes surgical videos to assess skill. Experts exhibit more efficient movements and use AI-driven fluctuation analysis for real-time skill evaluation.
Area of Science:
- Surgical Technology
- Medical Artificial Intelligence
- Laparoscopic Surgery
Background:
- Object detection technology using Artificial Intelligence (AI) enables precise tracking of surgical instruments in videos.
- Assessing surgical skill objectively is crucial for training and patient safety.
Purpose of the Study:
- To investigate kinematic differences in surgical instrument movements between expert and novice surgeons using AI.
- To evaluate the potential of AI-based kinematic analysis for real-time surgical skill assessment.
Main Methods:
- An AI algorithm was developed to accurately identify the X and Y coordinates of surgical instrument tips from video footage.
- Kinematic and fluctuation analyses were performed on 18 laparoscopic distal gastrectomy videos from three expert and three novice surgeons.
- Movement parameters including velocity, acceleration, jerk, and a fluctuation index (β) were analyzed.
Main Results:
- Expert surgeons demonstrated significantly more efficient and regular movements, characterized by reduced operation time and total travel distance.
- No significant differences in instrument tip velocity, acceleration, or jerk were observed between skill groups.
- The fluctuation index (β) was significantly higher in expert surgeons, with an ROC cutoff of 1.4 achieving 77.8% sensitivity and specificity.
Conclusions:
- AI-based object detection combined with fluctuation analysis shows promise for objective surgical skill evaluation.
- This method allows for real-time calculation of skill metrics, offering potential for peri-operative assessment.
- Further studies with larger sample sizes are warranted to validate these findings.
