Establish of an educational system for endoscopic surgical techniques with "quantification" and "self-feedback"

2023 Fiscal Year Final Research Report

• Project/Area Number: 21K18095
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 90140:Medical technology assessment-related
• Research Institution: Kyushu University
• Principal Investigator: Fukuta Atsuhisa 九州大学, 大学病院, 助教 (20709795)
• Project Period (FY): 2021-04-01 – 2024-03-31
• Keywords: 人工知能 / 小児外科 / 外科教育 / シミュレータ / 技術評価 / 外科

Outline of Final Research Achievements

We aimed to develop a new objective assessment system using artificial intelligence (AI) for forceps manipulation in a surgical training simulator.
To quantitatively evaluate tracking accuracy, we selected a random sample comprising 5% of the frames that were not used for AI training from the complete set of video frames. We compared the AI detection positions and correct positions and found an average pixel discrepancy of 9.2 pixels. To qualitative evaluate, the tracking stability was good at the forceps hinge; however, tracking of the forceps tip was unstable during rotation.
The AI-based forceps tracking system can visualize and evaluate endoscopic surgical skills. An objective evaluation not only facilitates the assessment of the surgeon’s achievements but also promotes effective learning. We believe that this system is a useful tool for the training and assessment of endoscopic surgeons.

Free Research Field

小児外科、外科教育

Academic Significance and Societal Importance of the Research Achievements

本研究によって内視鏡手術教育におけるAIの活用法を検証することができた。今回、着手したAIを活用した鉗子操作評価システムは検証を重ねることで更なる高度化が可能である。将来的にソフトウェア化することで，「いつでも」，「どこでも」利用可能な内視鏡外科医教育システムに繋がる有用なツールであると考えられる。AIを用いたフィードバックを行うことで、指導医の局在による教育格差をなくし、「質の高い医療を日本全体に整備する」という我が国の医療体制の下支えとしてのツールにつながる発展性をが期待される。