Lequan Yu

Papers

Total Citations

H-Index

About

Lequan Yu is a leading researcher at the forefront of computer vision and robotic surgery, with a primary focus on 3D reconstruction, deformable tissue modeling, and personalized federated learning. His most impactful work, *EndoGS: Deformable Endoscopic Tissues Reconstruction with Gaussian Splatting* (2025, 27 citations), introduces a groundbreaking approach that leverages Gaussian splatting to achieve rapid, high-fidelity 3D reconstruction of deformable tissues from single-viewpoint endoscopic videos. This method overcomes the time-consuming optimization of traditional dynamic radiance fields, offering a practical and efficient solution for real-time surgical visualization. Yu’s contributions extend to personalized federated learning for surgical instrument segmentation (SIS), as demonstrated in his 2025 paper (6 citations), where he enables multiple clinical sites to collaboratively train privacy-preserving models tailored to each site’s unique visual traits. This work addresses critical challenges in multi-site surgical data heterogeneity. With a total of 37 citations across his most cited works, Yu’s research is shaping the future of robotic surgery by enhancing both intraoperative tissue reconstruction and collaborative model personalization, making him a pivotal figure in advancing AI-driven surgical assistance.