Related Experiment Video
Updated: May 31, 2026

Reliability of Artificial Intelligence-Based Cone Beam Computed Tomography Integration with Digital Dental Images
Published on: February 23, 2024
Physical characterization of a CBCT scanner for maxillo-facial intra-surgical navigation
Clelia Cotena1, Carolina Mainardi2, Stefania Clemente3
1Dipartimento di Fisica "E. Pancini", Università di Napoli Federico II - Napoli, Italy.
Purpose:
To characterize the image quality and dose level of a cone-beam CT (CBCT) scanner dedicated to maxillo-facial intra-surgical navigation.
Methods:
Acquisition protocols for head imaging of average pediatric and adult patients were investigated. The detector response curve and the modulation transfer function (MTF) were assessed. The 3D image quality was evaluated using a commercial Catphan604 phantom in terms of accuracy of Hounsfield Unit (HU) estimates and slice uniformity. The MTF and the noise power spectrum (NPS) were assessed over the 3D reconstructed images. The weighted cone-beam dose index (CBDIw) was used for dosimetric benchmarking, and TLD chips placed in two anthropomorphic phantoms provided insight into the patient dose distribution.
Results:
The detector exhibited a response curve with a double natural logarithmic behavior and spatial resolutions up to 6.3 mm-1. The 3D spatial resolution ranged from 0.70 mm-1 at the scanner isocenter, for the pediatric protocol, to 0.46 mm-1 at 7 cm from the isocenter for the adult protocol. Due to beam hardening and the scatter, discrepancies between expected and measured HU were observed, in particular for positive HU values (i.e. those for acrylic, bone, delrin and teflon inserts). The CBDIw and dose distribution within anthropomorphic phantoms provided a dosimetric benchmark for the scanner.
Conclusion:
This study characterized a CBCT scanner in maxillo-facial surgery navigation. The findings support the definition of quality assurance protocols, providing methodologies and reference values for image quality and dose, while also highlighting limitations in spatial resolution and accuracy relevant for quantitative evaluations in clinical practice.
