Related Experiment Video
Updated: Jul 18, 2025
05:49
Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization
Published on: February 23, 2024
884
High-Accuracy 3D Image-Based Registration of Endoscopic Video to C-Arm Cone-Beam CT for Image-Guided Skull Base
Daniel J Mirota1, Ali Uneri1, Sebastian Schafer2
1Department of Computer Science, Johns Hopkins University, Baltimore MD.
Summary
A novel image-based method enhances endoscopic video registration to intraoperative imaging for high-precision head and skull-base surgery. This approach improves visualization and targeting accuracy compared to traditional tracker-based methods.
Area of Science:
- Medical Imaging
- Surgical Navigation
- Computer-Assisted Surgery
Background:
- Accurate registration of endoscopic video to preoperative CT is crucial for high-precision head, neck, and skull-base surgery.
- Conventional tracker-based registration methods have limitations in accuracy and do not leverage image data.
Purpose of the Study:
- To develop and evaluate a new image-based video registration method for improved accuracy in surgical navigation.
- To integrate intraoperative cone-beam CT (CBCT) for real-time anatomical updates and enhanced visualization.
Main Methods:
- Developed an image-based video registration method integrated with a navigation system using intraoperative C-arm cone-beam CT (CBCT).
- Employed an optical tracking system for initial endoscope localization, followed by direct 3D image-based registration to CBCT.
- Incorporated deformable registration of preoperative CT with intraoperative CBCT and 3D visualization.
Main Results:
- The image-based method achieved fast and accurate video-CBCT registration, enabling visualization of CBCT and planning data within the endoscopic video.
- Sub-millimeter accuracy (e.g., 0.92 mm TRE) was demonstrated in skull-base surgery applications, significantly outperforming conventional tracker-based registration (1.82 mm TRE).
- Consistent improvement in target registration error was observed in phantom and cadaver experiments.
Conclusions:
- The proposed image-based registration method offers a two-fold advance by utilizing up-to-date intraoperative CBCT and direct 3D image-based registration.
- This provides more confident visualization of critical structures and surgical targets within real-time imaging, leading to improved surgical precision.
