Related Experiment Video
Updated: Jun 4, 2026
09:10
Digital Hybrid Model Preparation for Virtual Planning of Reconstructive Dentoalveolar Surgical Procedures
Published on: August 5, 2021
CT-override: endoscopic updates to preoperative anatomical models during ablative surgery
Jan Emily Mangulabnan1, Jacob M Delgado López2, Lalithkumar Seenivasan2
1Johns Hopkins University, Baltimore, MD, 21211, USA. jmangul1@jh.edu.
Summary
This study introduces a novel method to dynamically update patient-specific surgical models using preoperative CT scans and intraoperative endoscopic video. This approach enhances surgical navigation accuracy by integrating visual cues with anatomical data.
Area of Science:
- Medical Imaging
- Computer-Aided Surgery
- Surgical Navigation
Background:
- Preoperative CT scans offer detailed anatomy for sinus surgery navigation but are static.
- Endoscope-based navigation reconstructs anatomy from visual cues, often lacking CT integration.
- Bridging pre- and intraoperative data is crucial for dynamic surgical scene updates.
Purpose of the Study:
- To develop a unified representation grounding both CT and endoscopic data.
- To enable dynamic updates of the surgical scene by integrating preoperative and intraoperative information.
- To improve surgical navigation by adapting to intraoperative tissue changes.
Main Methods:
- Initializing CT as a volumetric signed distance field (SDF).
- Reformulating updates with non-projective SDFs and accumulating camera-space observations.
- Estimating dense depth from endoscopic video for volume reconciliation and intraoperative SDF generation.
Main Results:
- Validation on cadaveric sinus surgery datasets with paired CT and endoscopic video.
- Extraction of intraoperative meshes at each surgical step, compared to ground-truth CT.
- Demonstrated submillimeter geometric accuracy and improved surface alignment of intraoperative representations.
Conclusions:
- Integrating CT priors with intraoperative vision enables anatomically consistent patient-specific model updates.
- The method provides a foundation for CT-consistent endoscopic reconstruction.
- Future work aims to integrate camera localization for a fully realized vision-guided navigation system.
