Related Experiment Video
Updated: Sep 26, 2026
Protocol for the Evaluation of MRI Artifacts Caused by Metal Implants to Assess the Suitability of Implants and the Vulnerability of Pulse Sequences
Published on: May 17, 2018
Effects of Dental Metal Implant Number and Configuration on PET/CT Quantification and Image Quality: A Phantom-Based
Emine Tuna Akdogan1,2, Sinem Coşkun2, Nami Yeyin2
1Department of Oral and Maxillofacial Surgery, Faculty of Dentistry, Istanbul Kent University, Istanbul Türkiye.
Introduction/Objectives:
Dental metal implants can generate artifacts in PET/CT imaging, particularly in the head and neck region, affecting quantitative accuracy. This study evaluated variations in standardized uptake value (SUV), contrast (C), and contrast-to-noise ratio (CNR) across predefined dental implant configurations differing in both implant number and spatial arrangement and assessed their effects on simulated tongue, nasopharyngeal, and laryngeal lesions.
Methods:
A custom-designed head and neck phantom containing extracted human teeth and up to eight titanium dental implants was used. Simulated tongue, nasopharyngeal, and laryngeal tumors were filled with F-18 FDG. PET/CT scans were acquired using seven implant configurations (0-8 implants) and reconstructed as attenuation-corrected (AC), AC with metal artifact reduction (AC+MAR), and non-attenuation-corrected (NoAC) images. SUV, contrast, and CNR were calculated using standardized ROIs.
Results:
Compared with the implant-free configuration, tongue tumor SUV decreased by 11% and 9.1% on AC and AC+MAR images, respectively, at eight implants. Nasopharyngeal and laryngeal SUV showed only minimal descriptive change with increasing implant number and configurations on AC+MAR images. Contrast and CNR followed similar trends on AC and AC+MAR images, whereas NoAC images consistently showed substantially lower contrast and CNR regardless of implant number.
Discussion:
Although MAR reduced implant-induced artifacts and partially attenuated SUV underestimation, it did not eliminate the configuration-dependent trend observed across the predefined implant arrangements. Lesions farther from the implants appeared less affected after MAR; however, this should be regarded as a hypothesis-generating observation from a single phantom experiment, requiring validation in patient studies and phantom models with varying implant-to-lesion distances.
Conclusion:
Dental implants may affect PET/CT quantification, particularly for lesions adjacent to multiple implants; SUV measurements in these regions should be interpreted with caution.
