Related Experiment Video
Updated: Jun 25, 2025
07:46
Author Spotlight: Revolutionizing Remote Surgery with Augmented Reality and Robotics for Enhanced Precision and Accessibility
Published on: August 9, 2024
705
Real-time deformable SLAM with geometrically adapted template for dynamic monocular laparoscopic scenes
Xuanshuang Tang1,2, Haisu Tao3,4, Yinling Qian5
1Department of Computer Science, Sichuan University, Chengdu, 610065, China.
Summary
This study introduces a new real-time deformable SLAM algorithm for accurate 3D reconstruction in surgery. The method uses a geometrically adapted template to improve capture of complex organ contours, enhancing surgical navigation systems.
Area of Science:
- Medical Imaging
- Computer Vision
- Robotics
Background:
- Intraoperative 3D reconstruction is vital for surgical navigation.
- Current deformable SLAM struggles with abrupt geometric features like organ contours.
Purpose of the Study:
- To develop a real-time monocular deformable SLAM algorithm for improved endoscopic scene reconstruction.
- To enhance the capture of abrupt geometric features in dynamic surgical environments.
Main Methods:
- A novel algorithm employs a geometrically adapted template for real-time monocular deformable SLAM.
- Dual-thread architecture (deformation mapping and tracking) ensures performance.
- Salient edge features guide template construction via triangulation.
Main Results:
- The proposed method demonstrates improved accuracy (0.75-7.95% enhancement) on benchmark datasets.
- Consistent effectiveness in data association compared to existing methods.
Conclusions:
- An adaptive template significantly improves reconstruction of dynamic scenes with geometric features.
- Further research is needed for incisal margins in laparoscopic surgery.
- This work advances computer-assisted navigation in laparoscopic surgery.
