Related Experiment Video
Updated: Aug 22, 2026
Reliability of Artificial Intelligence-Based Cone Beam Computed Tomography Integration with Digital Dental Images
Published on: February 23, 2024
Accuracy of CBCT and dental-MRI for detecting peri-implant bone defects at single titanium implants: an in-vitro
Katharina Hartung1, Nikolai Miotk1, Arne Lauer1
1Department of Neuroradiology, Heidelberg University Hospital, Im Neuenheimer Feld 400, 69120, Heidelberg, Germany.
Purpose:
To evaluate the reliability and accuracy of optimized dental-MRI (dMRI) for detecting and classifying peri-implant bone defects around single titanium implants.
Methods:
48 titanium implants were inserted into bovine ribs. 24 implants had surrounding standardized defects (1-wall, 2-wall, 3-wall, 4-wall) in two sizes (1 mm and 3 mm), while the other 24 implants showed no peri-implant bone defects. CBCT and dMRI were performed and analyzed twice by two readers. 3 Tesla MRI was conducted using an innovative intraoral coil and a novel, T1-weighted sequence for metal artifact reduction (CS-SEMAC). Cohen's Kappa (κ), sensitivity, and specificity were assessed for bone defect detection, size, and type. Modality differences were evaluated using the exact McNemar test and Firth penalized-likelihood logistic regression.
Results:
For bone defect detection, almost perfect intra-/inter-rater reliability (κ-value range: 0.958-1) was found for both imaging modalities, with high sensitivity/specificity (CBCT 100%/100%, 95% CI 0.962-1.000/0.962-1.000; MRI 98%/100%, 95% CI 0.927-0.998/0.962-1.000), confirmed by Firth regression (p = 0.22). For defect type classification, both CBCT (κ-value range: 0.884-0.913) and MRI (κ-value range: 0.794-0.942) demonstrated almost perfect intra-/inter-rater reliability. Overall sensitivity/ specificity were 86%/99% (95% CI 0.780-0.926 and 0.985-0.995) for CBCT and 84%/99% (95% CI 0.755-0.910 and 0.985-0.995) for dMRI, with no significant difference between modalities (Firth-adjusted OR 0.85, 95% CI 0.38-1.88, p = 0.68).
Conclusions:
Under optimized ex vivo conditions, dMRI offers a promising reliability and accuracy for identifying and classifying peri-implant bone defects around single titanium implants. High-resolution dMRI facilitates radiation-free, three-dimensional peri-implant tissue visualization.
