International Journal of Computerized Dentistry最新文献

Aim: The aim of the present study was to evaluate the effect of cement gap and drill offset on the marginal and internal fit discrepancies of crowns designed with different tooth preparations.

Materials and methods: Five tooth preparations were constructed, and crowns with different cement gaps and drill offsets were obtained. Then, best-fit alignment was performed on the crowns with the corresponding tooth preparations, and the fit discrepancies were expressed by color-coded difference images and root mean square (RMS) values. The RMS values of each group were analyzed by the rank-based Scheirer-Ray-Hare test (α = 0.05).

Results: The color segments in the sharp line angles area of the Sharp line angles group changed significantly before and after the drill offset. The cement gap had a significant effect on the marginal, internal, or overall fit discrepancies of the five design groups (P < 0.001), while the drill offset had a significant effect on the marginal fit discrepancies of the Shoulder-lip group and the internal or overall fit discrepancies of the Sharp line angles group (P < 0.001). Additionally, the interaction effect between cement gap and drill offset was significant for the marginal fit discrepancies of the Shoulder-lip group and the internal or overall fit discrepancies of the Sharp line angles group (P < 0.01).

Conclusions: The cement gap and drill offset had a significant adverse effect on the marginal or internal fit discrepancies of the crowns designed with the shoulder-lip and sharp line angles designs. Tooth preparation designs with intense curvature changes such as shoulder-lip and sharp line angles should be avoided clinically.

Aim: Although many studies in various fields employ deep learning models, only a few such studies exist in dental imaging. The present article aims to evaluate the effectiveness of convolutional neural network (CNN) algorithms for the detection and diagnosis of the quantitative level of dental restorations using panoramic radiographs by preparing a novel dataset.

Materials and methods: 20,973 panoramic radiographs were used, all labeled into five distinct categories by three dental experts. AlexNet, VGG-16, and variants of ResNet models were trained with the dataset and evaluated for the classification task. Additionally, 10-fold cross-validation (ie, 9 folds were separated for training and 1 fold for validation) and data augmentation were carried out for all experiments.

Results: The most successful result was shown by ResNet-101, with an accuracy of 92.7%. Its macro-average AUC was also the highest, at 0.989. Other accuracy results obtained for the dataset were 75.5% for AlexNet, 85.0% for VGG-16, 92.1% for ResNet-18, 91.7% for ResNet-50, and 92.1% for InceptionResNet-v2.

Conclusions: An accuracy of 92.7% is a very promising result for a computer-aided diagnostic system. This result proved that the system could assist dentists in providing supportive preliminary information from the moment a patient's first panoramic radiograph is taken. Furthermore, as the introduced dataset is powerful enough, it can be relabeled for different problems and used in different studies.

Aim: The present study aimed to evaluate the accuracy of automated detection of preparation finish lines in teeth with defective margins.

Materials and methods: An extracted first molar was prepared for a full veneer crown, and marginal defects were created and scanned (discontinuity of finish line: 0.5, 1.0, and 1.5 mm; additional line angle: connected, partially connected, and disconnected). Six virtual defect models were entered into CAD software and the preparation finish line was designated by 20 clinicians (CAD-experienced group: n = 10; CAD-inexperienced group: n = 10) using the automated finish line detection method. The accuracy of automatic detection was evaluated by calculating the 3D deviation of the registered finish line. The Kruskal-Wallis and Mann-Whitney U tests were used for between-group comparisons (α = 0.05).

Results: The deviation values of the registered finish lines were significantly different according to conditions with different amounts of finish line discontinuity (P < 0.001). There was no statistical difference in the deviation of the registered finish line between models with additional line angles around the margin. Moreover, no statistical difference was found in the results between CAD-experienced and CAD-inexperienced operators.

Conclusions: The accuracy of automated finish line detection for tooth preparation can differ when the finish line is discontinuous. The presence of an additional line angle around the preparation margin and prior experience in dental CAD software do not affect the accuracy of automated finish line detection.

Aim: To develop a deep learning (DL) artificial intelligence (AI) model for instance segmentation and tooth numbering on orthopantomograms (OPGs).

Materials and methods: Forty OPGs were manually annotated to lay down the ground truth for training two convolutional neural networks (CNNs): U-net and Faster RCNN. These algorithms were concurrently trained and validated on a dataset of 1280 teeth (40 OPGs) each. The U-net algorithm was trained on OPGs specifically annotated with polygons to label all 32 teeth via instance segmentation, allowing each tooth to be denoted as a separate entity from the surrounding structures. Simultaneously, teeth were also numbered according to the Fédération Dentaire Internationale (FDI) numbering system, using bounding boxes to train Faster RCNN. Consequently, both trained CNNs were combined to develop an AI model capable of segmenting and numbering all teeth on an OPG.

Results: The performance of the U-net algorithm was determined using various performance metrics including precision = 88.8%, accuracy = 88.2%, recall = 87.3%, F-1 score = 88%, dice index = 92.3%, and Intersection over Union (IoU) = 86.3%. The performance metrics of the Faster RCNN algorithm were determined using overlap accuracy = 30.2 bounding boxes (out of a possible of 32 boxes) and classifier accuracy of labels = 93.8%.

Conclusions: The instance segmentation and tooth numbering results of our trained AI model were close to the ground truth, indicating a promising future for their incorporation into clinical dental practice. The ability of an AI model to automatically identify teeth on OPGs will aid dentists with diagnosis and treatment planning, thus increasing efficiency.

Aim: To evaluate the accuracy and reproducibility of real and virtual occlusal contact points in implant-supported, fixed complete dentures.

Materials and methods: The study included 19 participants using mandibular interim complete-arch fixed prosthesis supported by 3 or 4 implants as opposed to conventional removable complete dentures. At installation, an examiner installed the prostheses and verified the occlusal contact points through 2 methods: recording the real contact points with carbon paper (RC) followed by occlusal photography and intraoral scanning (VC) to record the virtual contact points to obtain a screen print of the software. Then, the two images were randomized to determine the order to be inserted into Microsoft PowerPoint for blind and paired evaluation. The independent variables consisted of the distribution of occlusal contacts points (qualification through pre-defined scores based on the position of the contact points on the surfaces of the teeth) and the reproducibility of the methods by verifying the number of occlusal points. For this, a descriptive analysis was used to evaluate the distribution of occlusal contacts points and the Wilcoxon test for the reproducibility of the occlusal contact points between the methods (p<0.05).

Results: The methods had 100% and 73.6% real and virtual occlusal contact points, respectively, which is considered clinically excellent. There was no significant difference regarding the reproducibility of the methods by the number of occlusal contact points (RC: x̅13.32; VC: x̅13.68; p=0.715).

Conclusion: The use of the tested intraoral scanner can be an easy and fast tool for studying and mapping the occlusion, and storing data for future treatment, with the conventional method being the preferred method for performing the occlusal adjustment.

Aim: The aim of the present article is to describe a new method to reduce the undesirable loss of material thickness that results from overmilling due to the tool diameter compensation correction of common CAD/CAM software.

Materials and methods: Today's CAD/CAM software (eg, 3Shape or Exocad) specifies the same tool diameter compensation for different ceramics. In the case of zirconia ceramics milled in the raw state, this leads to excessive milling of the inner surfaces of crowns, which results in unnecessarily large cementation gaps and a restoration that is thinned out from the inside. By manually reducing the preset correction in the digital design process by the volumetric sintering shrinkage factor specified by the manufacturer, excessive thinning of the zirconia can be avoided.

Results: The inner geometry of the restorations changes only slightly after manually reducing the preset tool diameter compensation correction. Consequently, a design of the restoration with the required minimum interocclusal thickness yet with accurate passive seating and marginal fit is possible without any further interventions.

Conclusions: Understanding the specifics of the subtractive fabrication process as well as the properties of the restorative materials is a key factor in achieving optimal clinical outcomes with all-ceramic restorations fabricated with CAD/CAM technology. The use of monolithic zirconia combined with a calculated reduction in the preset tool diameter compensation correction might be beneficial in cases with thin or uneven geometry.

Aim: The aim of the present study was to compare the load bearing capacity of monolithic or bilayered single crowns with zirconia or polyetherketoneketone frameworks designed and fabricated using CAD/CAM systems.

Materials and methods: Cr-Co-based metal dies (n = 60) were duplicated from an extracted and prepared premolar and then restored with different CAD/CAM materials. The specimens were divided into five groups (n = 12) according to the type of materials: group S: monolithic zirconia-reinforced lithium disilicate; group ZI: bilayered zirconia framework with lithium disilicate; group ZE: bilayered zirconia framework with resin-infiltrated hybrid ceramic; group PI: bilayered polyetherketoneketone (PEKK) framework with lithium disilicate; group PE: bilayered PEKK framework with resin-infiltrated hybrid ceramic. The crown specimens were cemented on Cr-Co metal dies with a resin cement (Multilink N). The specimens were subjected to cyclic mechanical loading followed by load bearing testing. Data were analyzed using the Kruskal-Wallis and Mann-Whitney U tests (α = 0.05).

Results: The monolithic crowns in group S (1930 ± 452.18 N) presented significantly higher load bearing capacity than those of the other groups (P < 0.05). This was followed by group ZI (1165.41 ± 264.04 N). The remaining groups demonstrated comparable results. Failure types were more frequent in all zirconia specimens that included veneering ceramic.

Conclusions: Monolithic zirconia-reinforced lithium disilicate CAD/CAM ceramics showed superior load bearing capacity compared with all other bilayered counterparts. All CAD/CAM materials tested remarkably exceeded the average occlusal force in the posterior region.

Aim: To evaluate CAD/CAM milling vibrations related to zirconia disk design, framework, and location of digital crown nesting during zirconia soft milling.

Materials and methods: In the present in vitro study, 25 round zirconia disks of 98 mm (R group) and 15 D-shaped zirconia disks of 71 mm (D group) were placed in a 5-axis CAD/CAM milling unit. The R group had 3 crown milling sites (n = 75) and the D group had 2 crown milling sites (n = 30). In the R group, site A was located 57 mm, site B was 92 mm, and site C was 123 mm from the machine holder attachment. In the D group, site A was 57 mm and site C was 123 mm from the machine holder attachment. A vibration meter (VB-8200; Lutron) was connected to the disk holder, and data (mm/second) were collected during the milling process at the different nesting locations. One-way ANOVA with post-hoc tests and Bonferroni correction for multiple tests were used to compare the groups (α = 0.05).

Results: In the R group, site C exhibited the highest mean vibration values compared with site A (P = 0.001). The first and second cutting tools (2.5 and 1 mm) exhibited statistically significant differences between the D and R groups at milling sites A and C (P < 0.001 and P = 0.003, respectively).

Conclusions: Milling vibration increased as the distance between the machine holder arm and a digitally nested site increased. Round zirconia disks induced higher vibration values compared with D-shaped disks.