Related Experiment Video
Updated: May 11, 2026
3D Scanning Technology Bridging Microcircuits and Macroscale Brain Images in 3D Novel Embedding Overlapping Protocol
Published on: May 12, 2019
Effect of the CBCT data segmentation threshold on registration accuracy with surface scanning
Lucio Lo Russo1, Mauro Lorusso2, Carlo Ercoli3
1Professor of Prosthodontics, Department of Clinical and Experimental Medicine, School of Dentistry, University of Foggia, Foggia, Italy.
Statement Of Problem:
Accurate spatial relationships between cone beam computed tomography (CBCT) and intraoral surface scan (IOS) data is crucial to avoid encoding errors that can affect surgical and prosthetic planning, including surgical guide fabrication. Specifically, segmentation thresholds used during CBCT data processing may influence the geometry of 3D bone models and, consequently, the accuracy of their registration with the IOS data.
Purpose:
The purpose of this study was to assess whether different CBCT segmentation thresholds affected the registration accuracy with IOSs.
Material And Methods:
CBCT and IOS data from 19 patients were included in the study. Standard tessellation language (STL) models were generated from CBCT scans using 7 segmentation thresholds (250-1750 HU). IOSs were aligned onto each bone model, and deviations were measured by using a metrology software program along the X-, Y-, and Z-axes at 6 reference points. The effect of thresholds, arch type, tooth type, and side was analyzed using generalized estimated equation (GEE) models (α=.05).
Results:
A total of 231 models and 1306 measurements were obtained. The segmentation threshold significantly affected registration accuracy (P<.001 for all axes). The overall deviation ranged from -0.08 to 0.33 mm for the X-axis (anteroposterior direction), 0.01 to 0.03 mm for the Y-axis (left to right direction), and -0.13 to -0.66 mm for the Z-axis (apicocoronal direction). Measured deviations were consistent across arches and sides. In situations where a significant difference in Z-axis deviation was detected, irrespective of the segmentation threshold pair, the magnitude of this difference increased with higher segmentation thresholds.
Conclusions:
Segmentation threshold in CBCT data processing significantly impacted the registration of IOSs to STL bone models, particularly in the apicocoronal direction. This may result in errors transferred to subsequent surgical and prosthetic phases, including surgical guide fabrication, thereby affecting implant positioning. Clinicians should be aware of this source of inaccuracy and consider it during planning.
