Surgical instrument segmentation and classification in transcanal endoscopic ear surgery video using Segment Anything Model 2
This study demonstrated that SAM 2 can accurately identify surgical instruments in TEES videos without the need to prepare large, costly training datasets. This zero-shot segmentation characteristic significantly lowers the barrier to clinical implementation of AI, suggesting it is a promising foundational technology for future surgical navigation and robotic automation. Key future challenges include improving classification accuracy for articulating instruments and accelerating processing speed for practical, real-time use.
Authors
- Kazuhiro Matsui (ORCID: https://orcid.org/0000-0003-0634-2194)
- Keita Atsuumi
- Atsushi Nishikawa (ORCID: https://orcid.org/0000-0002-5852-9543)
- Toshikazu KAWAI (ORCID: https://orcid.org/0000-0001-6489-9543)
- Natsumi Uehara (ORCID: https://orcid.org/0000-0002-6620-4546)
- Hiroaki Hirai
- Toshihiko Yamashita (ORCID: https://orcid.org/0009-0000-1132-4455)
- Hisashi Suzuki (ORCID: https://orcid.org/0000-0002-7162-8730)
- Takeshi Fujita (ORCID: https://orcid.org/0000-0002-6457-6231)
- R Ueno
Institutions
- Toyo University (JP)
- Osaka Prefectural Toyonaka Support School (JP)
- Hiroshima City University (JP)
- Osaka Institute of Technology (JP)
- Chuo University (JP)
Publication Details
- Journal
- Quantitative Imaging in Medicine and Surgery
- Published
- 2025-11-25
- DOI
- https://doi.org/10.21037/qims-2025-1469
- Primary Topic
- Soft Robotics and Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.00