The aim was to assess the precision and accuracy of cephalometric analyses performed by artificial intelligence (AI) with and without human augmentation. Four dental professionals with varying experience levels identified 31 landmarks on 30 cephalometric radiographs twice. These landmarks were re-identified by all examiners with the aid of AI. Precision and accuracy were assessed by using intraclass correlation coefficients (ICCs) and mean absolute errors (MAEs). AI revealed the highest precision, with a mean ICC of 0.97, while the dental student had the lowest (mean ICC: 0.77). The AI/human augmentation method significantly improved the precision of the orthodontist, resident, dentist, and dental student by 3.26%, 2.17%, 19.75%, and 23.38%, respectively. The orthodontist demonstrated the highest accuracy with an MAE of 1.57 mm/°. The AI/human augmentation method improved the accuracy of the orthodontist, resident, dentist, and dental student by 12.74%, 19.10%, 35.69%, and 33.96%, respectively. AI demonstrated excellent precision and good accuracy in automated cephalometric analysis. The precision and accuracy of the examiners with the aid of AI improved by 10.47% and 27.27%, respectively. The AI/human augmentation method significantly improved the precision and accuracy of less experienced dental professionals to the level of an experienced orthodontist.

The aim of this study was to compare cephalometric measurements obtained from conventional cephalograms with total and half-skull synthesized cone beam CT (CBCT) cephalograms. Cephalometric analyses of 30 clinically symmetric patients were conducted by a calibrated examiner on conventional and CBCT-synthesized cephalograms (total, right and left). Reproducibility was investigated using the intraclass correlation coefficient (ICC). The Bland-Altman analysis was used to assess the agreement of the measurements from each factor obtained by conventional, total, right and left CBCT-synthesized cephalograms. The ICC was above 0.9 for most of the 40 cephalometric factors analysed, revealing similar levels of reproducibility. When the measurements obtained from conventional and CBCT-synthesized cephalograms were compared, the Bland-Altman analysis showed a strong agreement between them. Half-skull CBCT-synthesized cephalograms offer the same diagnostic performance and equivalent reproducibility in terms of cephalometric analysis as observed in conventional and total CBCT-synthesized cephalograms.

Purpose: To investigate the trueness of intraoral scanning in 8 commonly seen partially edentulous conditions.Materials and methods: A maxillary dentoform was modified into the 8 commonly seen partially edentulous conditions. Each modification was scanned with a laboratory desktop scanner. Each modification was then scanned 10 times (n = 10) with an intraoral scanner. All scans were exported as STL files and then imported into a surface matching software using the best‐fit alignment method. The dimensional differences between the study STL files from the intraoral scanner were compared to the corresponding reference STL files. The measurements were calculated as the root mean square (RMS) and defined as the trueness of the intraoral scans. In addition to the RMS values, qualitative assessments were completed on the color maps. The color maps produced by the surface matching software were used to visualize the areas of deviation between scans from the intraoral scanner and their corresponding reference files. One‐way analysis of variance (ANOVA), followed by pair‐wise comparisons using Fisher's Protected Least Significant Difference were utilized to compare the differences between the groups in RMS values (α = 0.05).Results: Partially edentulous condition significantly affected the trueness of the intraoral scans. Group 8 (Class IV) had significantly lower RMS (0.1878 ± 0.0455 mm) than all other groups (p < 0.001). Group 2 (Class II) and Group 7 (Class III modification I) are not significantly different from each other (Group 2: 0.5758 ± 0.0300 mm; Group 7: 0.5602 ± 0.0231 mm, p = 0.571), while they both had significantly higher RMS than all other groups (p < 0.001). The remaining groups showed the RMS values were within the range of 0.3001 ± 0.0891 mm (Group 6 – Class III with Long Edentulous Span) and 0.4541 ± 0.1039 mm (Group 1 – Class I).Conclusion: Different partially edentulous conditions affected the trueness of the scans generated from the selected intraoral scanner. Class IV edentulous condition had the highest intraoral scan trueness. It is unknown if RMS values are clinically significant, and the validity of using intraoral scans directly for PRDP fabrication will need further studies.