Surgical Instrument
Surgical instrument research focuses on developing advanced computer vision and robotic systems to improve surgical precision, efficiency, and safety. Current efforts concentrate on real-time 3D pose estimation and segmentation of instruments using deep learning models, such as neural networks (including YOLO and Transformers) and neural fields, often trained on synthetic data to overcome limitations in annotated real-world datasets. These advancements enable applications like augmented reality guidance, robotic teleoperation, and improved surgical workflow analysis, ultimately contributing to better patient outcomes and more efficient surgical procedures.
Papers
Monocular pose estimation of articulated surgical instruments in open surgery
Robert Spektor, Tom Friedman, Itay Or, Gil Bolotin, Shlomi Laufer
SegSTRONG-C: Segmenting Surgical Tools Robustly On Non-adversarial Generated Corruptions -- An EndoVis'24 Challenge
Hao Ding, Tuxun Lu, Yuqian Zhang, Ruixing Liang, Hongchao Shu, Lalithkumar Seenivasan, Yonghao Long, Qi Dou, Cong Gao, Mathias Unberath