Human examination and artificial intelligence in cephalometric landmark detection-is AI ready to take over?

OBJECTIVES: To compare the precision of two cephalometric landmark identification methods, namely a computer-assisted human examination software and an artificial intelligence program, based on South African data. METHODS: This retrospective quantitative cross-sectional analytical study utilized a data set consisting of 409 cephalograms obtained from a South African population. 19 landmarks were identified in each of the 409 cephalograms by the primary researcher using the two programs [(409 cephalograms x 19 landmarks) x 2 methods = 15,542 landmarks)]. Each landmark generated two coordinate values (x, y), making a total of 31,084 landmarks. Euclidean distances between corresponding pairs of observations was calculated. Precision was determined by using the standard deviation and standard error of the mean. RESULTS: The primary researcher acted as the gold-standard and was calibrated prior to data collection. The inter- and intrareliability tests yielded acceptable results. Variations were present in several landmarks between the two approaches; however, they were statistically insignificant. The computer-assisted examination software was very sensitive to several variables. Several incidental findings were also discovered. Attempts were made to draw valid comparisons and conclusions. CONCLUSIONS: There was no significant difference between the two programs regarding the precision of landmark detection. The present study provides a basis to: (1) support the use of automatic landmark detection to be within the range of computer-assisted examination software and (2) determine the learning data required to develop AI systems within an African context.

Custom Focal Trough in Cone-Beam Computed Tomography Reformatted Panoramic Versus Digital Panoramic for Mental Foramen Position to Aid Implant Planning.

OBJECTIVES: To compare the linear measurements from digital panoramic (DP) radiographs and cone-beam computed tomography (CBCT) volumes for the localization of the mental foramen (MF). MATERIAL AND METHODS: Thirty-one patients with panoramic and CBCT radiographs depicted on the same machine were analyzed. The vertical and horizontal positions of the MF were compared by the differences in distances measured from reference points to the boundaries (tangents) of the MF in digital panoramic (DP) and CBCT reformatted panoramic (CRP) views. The vertical position of MF was also analyzed on CBCT oblique coronal views (CORO) and compared with its corresponding distances on DP and CRP views. RESULTS: Statistically significant differences (P < 0.05) were found in all compared measurements between CRP and DP views. In addition, the vertical distance (Y1) compared between DP, CRP, and CORO views also showed a statistically significant measurement discrepancy in the mean distance (P < 0.000) with the highest mean difference of 1.59 mm (P < 0.05) was attained from Y1 (DP-CORO). Inter- and intra-examiner analysis indicated a high level of agreement for all measurements. CONCLUSION: The mean values of discrepancies in measurements between DP and CRP views for horizontal and vertical linear measurements were clinically tolerable. Nevertheless, significant differences in the vertical MF position were detected between the panoramic views (DP, CRP) and the coronal views (CORO). This implies that the use of coronal view measurements during implant planning might reduce the risk of neurovascular injuries.