Papers
12
Total Citations
143
H-Index
6
About
An Wang is a leading researcher at the intersection of computer vision and robotic surgery, whose work is driving the next generation of intelligent surgical systems. His primary research areas include surgical instrument segmentation, action recognition, and robust visual question-answering for robot-assisted procedures. Wang’s major contributions lie in rigorously testing and adapting foundation models—most notably the Segment Anything Model (SAM)—for the unique challenges of the operating room. His empirical studies on SAM’s generalization and robustness in robotic surgery (accumulating over 50 citations) have been pivotal in revealing how these powerful models can be made reliable for clinical use. He has also advanced weakly supervised learning for object detection in challenging domains like underwater imaging and introduced adversarial contrastive learning for calibrated, robust visual question-localized answering (Surgical-VQLA++). Wang’s impact is further evidenced by his involvement in the SAR-RARP50 challenge and the Intuitive Surgical SurgToolLoc and SurgVU challenges, where he helps benchmark and drive progress in surgical data science. His work is not only pushing the boundaries of AI but also laying the groundwork for safer, more autonomous robotic surgery.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3Proposal-Refined Weakly Supervised Object Detection in Underwater Images18 citations · 2019
- 4
- 5SAM Meets Robotic Surgery: An Empirical Study in Robustness Perspective12 citations · 2023
- 6
- 7
- 8Intuitive Surgical SurgToolLoc and SurgVU Challenges Results: 2022-20256 citations · 2023
- 9
- 10