Chairs:
reinier hoogeveen
tomasz kulczyk
76: ACCURACY OF MRI-DERIVED VIRTUAL 3D BONE MODELS & CLINICAL FEASIBILITY IN MRI-BASED GUIDED IMPLANT SURGERY
F. Probst1, Y. Malenova2, D. Karampinos3, M.J. Stumbaum4, E. Burian3, J. Schweiger4, M. Probst5
1Ludwig-Maximilians-University München, Oral and Maxillofacial Surgery and Facial Plastic Surgery, München, Germany, 2Ludwig-Maximilians-University Munich, Oral and Maxillofacial Surgery and Facial Plastic Surgery, Munich, Germany, 3Technical University of Munich, Department of Diagnostic and Interventional Radiology, Munich, Germany, 4Ludwig-Maximilians-University Munich, Department of Prosthetic Dentistry, Munich, Germany, 5Technical University of Munich, Department of Diagnostic and Interventional Neuroradiology, Munich, Germany
Aim: To investigate the accuracy of 3D bone surface models derived from MRI against CT and CBCT benchmarks, and to assess if MRI-based guided implant surgery is feasible in a clinical pilot study.
Material and Methods: The first study utilized porcine mandibles, comparing MRI, CT, and CBCT derived models against optical scans (reference) for geometric accuracy. The second study involved a clinical pilot with 12 patients undergoing MRI-based virtual planning and guided implant surgery, assessing deviations between planned and actual implant positions.
Results: MRI-derived 3D bone models demonstrated comparable geometric accuracy to CT and CBCT. All modalities were within a ± 0.3 mm equivalence margin, confirming their consistency and reliability. The clinical implementation of MRI-based guided implant surgery indicated acceptable deviations between planned and resulting implant positions (entry point error 0.8 ± 0.3 mm, apex error 1.2 ± 0.6 mm, angular deviation 4.9 ± 3.6°), as well as effective visualization of relevant anatomical structures.
Conclusion: MRI technology presents a reliable alternative to CT and CBCT for computer-assisted maxillofacial surgery, and MRI-based guided implant surgery is a feasible and accurate procedure that avoids exposure to ionizing radiation.
87: DENTAL-DEDICATED MAGNETIC RESONANCE IMAGING AND CONE BEAM COMPUTED TOMOGRAPHY FOR APICAL PERIODONTITIS ASSESSMENT: A COMPARATIVE IN VIVO PILOT STUDY
J. Fuglsig1, J. Christensen1, K. Johannsen1, R. Spin-Neto1
1Aarhus University, Department of Dentistry and Oral Health, Aarhus C, Denmark
Aim: To compare dental-dedicated magnetic resonance imaging (ddMRI) to cone beam CT (CBCT) as an imaging modality for diagnosing apical periodontitis (AP).
Material and Methods: Ten patients referred for CBCT imaging (X1, 3Shape, Copenhagen, Denmark; field-of-view 5×5 cm, resolution 0.15 mm) due to suspicion of a periapical lesion were also imaged using ddMRI. ddMR images were acquired on a Magnetom Free.Max (Siemens Healthineers, Erlangen, Germany) equipped with a dental-dedicated coil (Rapid-Biomedical, Rimpar, Germany), using two sequences (6×6 cm FOV, section thickness 2 mm): proton-density-weighted turbo spin-echo (PD-TSE) for sagittal and coronal sections (resolution 0.3 mm) and a PD-TSE with fat suppression for sagittal sections (resolution 0.4 mm). Three observers assessed the presence of AP (disease/healthy/unsure) in representative sections of the CBCT and ddMRI. Inter- and intra-observer reproducibility were assessed by means of Kappa statistics. A mathematical consensus observer was defined for each of the two modalities based on the scores from the three observers, and these were statistically compared (McNemar test).
Results: Inter-observer reproducibility was moderate (0.61-0.8) for CBCT, and ranged from moderate to excellent for ddMRI (0.62-1.0). Intra-observer reproducibility ranged from moderate to excellent for CBCT (0.61-1.0) and was moderate (0.66-0.79) for ddMRI. There was no statistically significant difference between CBCT and ddMRI (p=0.94).
Conclusion: In this pilot study, ddMRI was comparable to CBCT regarding the assessment of apical periodontitis in humans. Further education and specific ddMRI training is needed to improve observers’ expertise with this novel imaging modality.
110: ACCURACY OF DENTAL MRI FOR FULL-NAVIGATED DENTAL IMPLANT PLANNING IN A PRIVATE PRACTICE SETTING
C. Ott1, O. Möbes2, A. Skenderi1, S. Heiland1, M. Bendszus1, T. Hilgenfeld1
1Heidelberg University Hospital, Department of Neuroradiology, Heidelberg, Germany, 2Dental practice, Ettenheim, Germany
Aim: To investigate the accuracy of high-resolution dental MRI (dMRI) for full-navigated dental implant planning in a private practice setting.
Material and Methods: 13 patients scheduled for dental implant placement underwent 3T dMRI utilizing a 0.4-mm isotropic proton density (PD) weighted sequence within a 10-minute protocol. Subsequently, utilizing dedicated guided surgery software, virtual implant planning and surgical guide printing was executed. Surgical guides were placed intraorally during subsequent Orthopantomography. Finally implant placements were performed using a full-navigated protocol. The 3D position accuracy of 15 dental implants was evaluated by comparing the virtually planned and definitive implant position. The latter was defined by an intraoral scan.
Results: All dMRI-derived implant plans were suitable for implant planning. Three-dimensional deviations noted were 0.88 ± 0.69 mm at the entry point and 1.59 ± 0.78 mm at the apex, along with a mean angular deviation of 4.65 ± 2.79 degrees.
Conclusion: Within the limitations of this small patient cohort, dMRI-derived implant planning was feasible as all implant plans were suitable for surgery and observed accuracy was within the published range of CBCT-derived implant planning accuracy.
115: DEVELOPMENT AND TESTING OF A CARRIER PHANTOM FOR EXAMINING DENTAL MATERIALS IN DENTAL-DEDICATED MAGNETIC RESONANCE IMAGING
M. Sampaio-Oliveira1,2, M. L. Oliveira1, R. Spin-Neto2
1Piracicaba Dental School, University of Campinas, Oral Diagnosis, Piracicaba, Brazil, 2Aarhus University, Dentistry and Oral Health, Aarhus, Denmark
Aim: To develop and test a phantom to be used as a carrier for examining dental materials in dental-dedicated magnetic resonance imaging (ddMRI).
Material and Methods: The phantom was composed of five glass test tubes vertically positioned in a glass beaker. All tubes were filled with air, distilled water, 1.5% agar, 0.3 g.L-1 Ni(NO3)2 in 1.5% agar, or 1000 g.L-1 K2HPO4. The beaker was filled with distilled water, 0.3 g.L-1 Ni(NO3)2 aqueous solution, or 1000 g.L-1 K2HPO4 aqueous solution. The phantom was scanned using a ddMRI system with a dental-dedicated radiofrequency surface coil and a proton density turbo-spin-echo pulse sequence. Triplicate scans were conducted for each combination of five tube fillings with three beaker solutions, yielding 45 image volumes. Mean voxel value (MVV), voxel value inhomogeneity (VVI), image noise (IN), and contrast were evaluated.
Results: For the tubes, MVV ranged from 9.19 (air) to 322.31 (Ni(NO3)2, with water). VVI ranged from 1.59 (air, with water) to 24.74 (K2HPO4). IN varied from 5.46 (air, with water) to 16.26 (Ni(NO3)2 in agar). For the phantom body, MVV ranged from 196.54 (water, with air) to 391.99 (Ni(NO3)2, with Ni(NO3)2 in agar). VVI ranged from 10.50 (Ni(NO3)2, with K2HPO4) to 18.62 (Ni(NO3)2, with agar). IN ranged from 5.23 (water, with air) to 18.41 (K2HPO4). Contrast ranged from 12.67 (water) to 362.25 (air, with Ni(NO3)2).
Conclusion: The developed phantom composed of tubes filled with 1.5% agar surrounded by a 1000 g.L-1 K2HPO4 aqueous solution seems to be feasible for ddMRI quality assessments.
116: IMAGE ARTIFACTS GENERATED BY ZIRCONIA IMPLANTS AND THEIR IMPACT ON MAGNETIC RESONANCE IMAGE QUALITY
J. Schmitz1, L.-A. Schaafs2, C. Kulisch3, S. Nahles1, M. Heiland1, T. Flügge1
1Charité Universitätsmedizin Berlin, Department of Oral and Maxillofacial Surgery, Berlin, Germany, 2Charité Universitätsmedizin Berlin, Department of Radiology, Berlin, Germany, 3Charité Universitätsmedizin Berlin, Institute of Functional Anatomy, Berlin, Germany
Aim: This study assessed the artifact formation in MR scans caused by dental implants ex vivo in human mandibles. It was hypothesized that zirconia implants generate fewer artifacts than titanium implants and, therefore, the surrounding anatomy can be accurately assessed.
Material and Methods: Four dental implants with varying diameters, heights, and materials combined with different abutments and crowns were used. The implants were placed in partially edentulous human mandibles. MRI at 3.0 Tesla and CBCT were acquired of each mandible before and after implant placement. The extent of artifacts generated by each implant was assessed on 3D T1-weighted MRI. The mean size of the artifact was divided by the actual size of each implant to show overestimation. To qualitatively evaluate the images using CBCT images as reference, buccal and oral bone thickness at the implant shoulder and the distance between the implant tip and nerve canal were measured. Variances were calculated using one-way ANOVA with Bonferroni correction.
Results: The overestimation of titanium implants in diameter and length on MR images was significantly higher compared to zirconia implants (p<0.001). The underestimation of buccal and lingual bone crest in MR images with titanium implants was also significant compared to zirconia implants (p<0.001). No significant difference between titanium and zirconia implants in underestimating the distance implant tip–nerve canal was found.
Conclusion: Zirconia implants generate significantly fewer MRI artifacts compared to the titanium implant. Those artifacts impair the assessment of periprosthetic anatomy. The artifact size was not influenced by abutment and crown material.