Related Experiment Video
Updated: Jul 7, 2026
09:42
Murine Endoscopy for In Vivo Multimodal Imaging of Carcinogenesis and Assessment of Intestinal Wound Healing and Inflammation
Published on: August 26, 2014
Anatomical reconstruction from endoscopic images: toward quantitative endoscopy.
Hanzi Wang1, Daniel Mirota, Gregory Hager
1Center for Computer-Integrated Surgical Systems and Technology, The Johns Hopkins University, Baltimore, Maryland 21218, USA.
American Journal of Rhinology
|February 21, 2008
Summary
This study demonstrates the feasibility of creating detailed 3D anatomic models from endoscopic images using computational vision. These reconstructions offer significant potential for improving surgical navigation and treatment planning.
Area of Science:
- Computational imaging
- Medical visualization
- Surgical technology
Background:
- Recent computational imaging advances enable camera motion and scene geometry reconstruction from monocular images.
- Applying these techniques to endoscopic sequences allows for detailed, quantitative anatomic reconstructions.
- These reconstructions hold significant potential for various clinical applications.
Purpose of the Study:
- To develop a process flow for reconstructing anatomy from endoscopic image sequences.
- To validate the feasibility of computing such anatomic reconstructions.
Main Methods:
- Outlined a comprehensive process flow for endoscopic reconstruction.
- Implemented the process using advanced computational vision methods.
- Validated the approach on cadaverous specimens with known ground truth motion.
Main Results:
- Achieved consistent estimation of endoscopic motion.
- Generated dense anatomic reconstructions for over 65% of image pairs.
- Successfully processed 1373 image pairs.
Conclusions:
- Processing endoscopic images for anatomic structure reconstruction is feasible.
- These reconstructions show high potential clinical value.
- Applications include intraoperative navigation, diagnosis, and treatment planning.
