Papers
6
Total Citations
124
H-Index
5
About
Yu Takeda is a leading researcher at the intersection of computer vision, deep learning, and robotic-assisted minimally invasive surgery (MIS). His work focuses on overcoming key visualization and autonomy challenges in knee arthroscopy, where surgeons lose direct sight of the operative field. Takeda’s major contributions include pioneering the use of deep learning for automatic segmentation of multiple anatomical structures in arthroscopic scenes, achieving 41 citations for his foundational 2020 work. He also developed supervised scene illumination control for stereo arthroscopes (28 citations) and introduced Bayesian CNNs to quantify segmentation uncertainty in 4D ultrasound images of femoral cartilage (20 citations), a critical step for safe robotic guidance. Notably, Takeda proved the feasibility of using high-refresh-rate 4D ultrasound to create a volumetric knee joint atlas for autonomous robotic platforms (15 citations). His recent work on surface reflectance metrics for untextured surgical scene segmentation (2023) further advances real-time intraoperative perception. With over 120 total citations, Takeda’s research is shaping the future of autonomous, image-guided MIS, directly impacting patient safety and surgical precision.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6Surface Reflectance: A Metric for Untextured Surgical Scene Segmentation5 citations · 2023