Related Experiment Video
Updated: Sep 2, 2025

11:28
Robotic-Guided Stereoelectroencephalography for Invasive Epilepsy Monitoring
Published on: June 13, 2025
520
StereoCNC: A Stereovision-guided Robotic Laser System.
Guangshen Ma1, Weston Ross1,2, Patrick J Codd1,2
1Brain Tool Lab. Department of Mechanical Engineering, Duke University.
Summary
This study introduces StereoCNC, a robotic laser surgery system using stereovision for precise targeting. It achieves high accuracy (0.13mm RMSE) for superficial laser ablation, benefiting dermatology and neurosurgery.
Area of Science:
- Robotics
- Computer Vision
- Surgical Technology
Background:
- Robotic laser surgery systems require precise targeting for effective ablation.
- Integrating stereovision enhances depth perception and accuracy in robotic systems.
Purpose of the Study:
- To develop and validate an end-to-end stereovision-guided laser surgery system (StereoCNC) for precise laser ablation.
- To improve targeting accuracy by incorporating a 3D error field and optimization algorithms.
Main Methods:
- Integration of two digital cameras into a robotic laser system for stereovision.
- 3D-to-3D least-squares calibration to register camera and laser coordinate frames.
- Gaussian Process Regression (GPR) to model a 3D error field based on calibration reprojection errors.
- Genetic Algorithm optimization to guide mechanical stages for precise surgical site positioning within the error field.
Main Results:
- The StereoCNC system achieved an average Root Mean Square Error (RMSE) of 0.13 ± 0.02 mm.
- Maximum targeting error was measured at 0.34 ± 0.06 mm.
- Experiments on phantoms demonstrated the system's effectiveness in targeting various shapes and textures.
Conclusions:
- The developed stereovision-guided robotic system offers high precision for superficial laser surgery.
- Potential applications include dermatologic procedures and removal of exposed tumorous tissue in neurosurgery.

