Comparing the precision of reproducibility of computer-aided occlusal design to conventional methods

Automatic adjustment of dental crowns using Laplacian mesh editing

Currently, the restoration of missing teeth by means of dental implants is a common treatment method in dentistry. Ensuring optimal contact between teeth (occlusion) when designing the occlusal surface of an implant-supported crown is crucial for the patient. Although there are various occlusal concepts and guidelines for achieving optimised occlusion, adapting an occlusal surface is challenging. The contact points must be established in certain areas of the occlusal surface without impairing the aesthetics of the teeth and the masticatory function. A computer-aided, automated modelling approach can assist in the design process and can reduce the reliance on manual labour. This study aimed to develop a modelling approach that enables the automatic adaptation of an occlusal surface to specific occlusal concepts while preserving the natural appearance. In this study, the occlusal surface of an implant-supported crown based on a scanned first right mandibular molar was adopted. Nominal contact points were determined based on occlusal concepts by Ramfjord and Ash (RA) and Thomas (T). The shape of the occlusal surface was then adapted concerning the desired contact points using Laplacian mesh editing. The modification results were validated for different forces and crown materials (3Y-TZP and PMMA) using a finite element contact analysis. The contact analysis results showed that locations with high compressive stresses correspond with the locations of the nominal contact points. The reaction forces were more evenly distributed in PMMA crowns, due to the lower Young's modulus of PMMA compared to 3Y-TZP. Furthermore, the occlusal scheme with fewer contact points (RA) showed higher maximum reaction forces per contact area. The presented method enables the automated adaptation of an (implant-supported) crown to specific occlusal schemes, proving to be valuable in dental CAD.

Feasibility and accuracy of single maxillary molar designed by an implicit neural network (INN)-based model: A comparative study

PURPOSE

This feasibility study aimed to develop and evaluate a novel implicit neural network (INN)-based model for the automatic reconstruction of biomimetic maxillary molar.

MATERIAL AND METHODS

A total of 500 sets of full dental scans containing intact right and left maxillary first molars (#3 and #14) and adjacent teeth were included in this study. The digital maxillary casts were duplicated: one set served as the original, while the other had one maxillary first molar removed. Two INN-based models were developed, including the point convolution for surface reconstruction (POCO-only) model and the POCO with point multilayer perceptron (POCO-PointMLP) model. Each model was trained with either 12,000 or 50,000 sampling points from the training dataset. Chamfer distance (CD), F-score, and Volumetric Intersection over Union (volumetric IoU) were used to evaluate the INN-based models and training outcomes. The best INN-based model was selected to generate 20 digital crown designs (GCs) automatically from randomly selected test datasets and compared with the digital crowns designed by experienced human laboratory technicians (TCs). Both GCs and TCs were compared with the original clinical crowns (OCs) using root mean square (RMS) and color maps. Comparisons of RMS values were analyzed utilizing paired t-tests with statistical significance set at α = 0.05.

RESULTS

The performance of 2 INN-based models and training outcomes were evaluated and POCO-PointMLP trained with 50,000 sample points showed the best outcomes, with CD of 0.000119403, F-score of 0.919740566, and volumetric IoU of 0.943382978. This INN-based model was then selected to generate 20 digital crown designs (GCs) and compared with TCs for the RMS outcomes. The GCs exhibited numerically lower mean RMS compared to TCs without significant difference (0.2839 ± 0.0307 versus 0.3026 ± 0.0587 mm, p = 0.202). GCs and TCs both closely matched to the original clinical crowns (OCs).

CONCLUSIONS

This study demonstrated the feasibility and accuracy of the proposed INN-based POCO-PointMLP model for automating the digital crown designs of maxillary first molars. The INN-based model-generated digital crown designs showed comparable 3D deviations to the ones designed by experienced human laboratory technicians, presenting a novel and efficient artificial neural network architecture for tooth digital design tasks with the expansion of training datasets, refined training hyperparameters, and integrated comprehensive assessments.

Comparison of the morphological accuracy of automatic crowns designed by multiple computer-aided design software programs with different levels of dentition information acquisition

STATEMENT OF PROBLEM

Information on the morphological accuracy of crowns automatically produced by different computer-aided design (CAD) software programs for multilevel dentition defects is limited.

PURPOSE

The purpose of this in vitro study was to compare the morphological accuracy of crowns fabricated using different CAD software programs with different design theories for multilevel dentition defects.

MATERIAL AND METHODS

Four dentition defect types (the standard group, the abrasion group, the adjacent-teeth-missing group, and the antagonist-tooth-missing group, n=10) were fabricated to represent different levels of missing dentition information. Two design modes (the library mode and correlation mode) of 3 common CAD software programs (3Shape [3Shape A/S], CEREC [Dentsply Sirona], and exocad DentalCAD [exocad GmbH]) were used to design crowns automatically, and the morphologies of the generated crowns and original teeth were recorded. The root mean square (RMS) value was calculated to evaluate the morphological deviations between the autogenerated crowns and original teeth using the 3D matching system (Geomagic GmbH). As each group in this study represented 3 factors, the mean differences between the treatment combinations and the interaction effects were analyzed by performing factorial analysis of variance (α=.05).

RESULTS

The RMS values of autogenerated crowns designed using the correlation method were significantly lower than those designed using the library method of each software program in the 4 groups (P<.05). The RMS values of crowns designed by the 3Shape and CEREC software programs in library mode under conditions with dentition information loss were lower than those of crowns designed by the exocad software program (P<.05). Changes in the acquisition of dental information did not decrease the CEREC design accuracy (P>.05), while they did decrease the 3Shape and exocad design accuracy (P<.05).

CONCLUSIONS

The correlation method showed higher accuracy in rebuilding the original morphology of the teeth than the library method. Both the 3Shape and CEREC software programs showed higher accuracy than the exocad software program in library mode under conditions with dentition information loss, while CEREC showed higher stability than the 3Shape and exocad software programs.

A digital technique for occlusal design of a posterior full crown based on physiological tooth displacement and occlusion of adjacent teeth

This article presents a digital technique to construct a virtual occlusion in the maximal intercuspal position (MIP), considering physiological tooth displacement and reducing intermesh penetrations between occlusal surfaces, in order to design more precise and accurate occlusal contacts of a posterior full crown.

Occlusal contact and clearance of posterior implant-supported single crowns designed by two different methods: a self-controlled study

BACKGROUND

Precise occlusal design of implant-supported fixed prostheses is difficult to achieve by the conventional wax-up method, often requiring chairside adjustments. The computer-aided design (CAD) method is promising. This study aims to compare the occlusal contacts and clearance of posterior implant-supported single crowns designed by the CAD and conventional methods.

METHODS

Sample size calculation indicated fourteen samples per group. Two sets of type-IV plaster casts with a single implant analog inserted in the posterior teeth region were mounted as master casts in a mechanical articulator in maximal intercuspal position (MIP). Seven working cast sets were obtained from each master cast by a closed tray technique, and mounted in MIP. Two implant-supported single crowns were designed with an occlusal clearance to achieve light occlusal contact in each working cast set by CAD and conventional method, separately. For the CAD group, the crown was designed in digital models obtained by scanning the working casts. For the conventional group, wax-up of the crown was prepared on the working casts and scanned to generate a STL file. In the working and master casts, mean and minimum occlusal clearances in the designed occlusal contact area of the both finished prostheses were calculated using the occlusal clearance (OC) and occlusal record (OR) method. The prostheses' occlusion was evaluated in master casts.

RESULTS

For the evaluation in the working casts, both design methods had similar mean occlusal clearances by the OC method (195.4 ± 43.8 vs. 179.8 ± 41.8 μm; P = 0.300), while CAD group resulted in a significantly larger minimum occlusal clearance in the designed occlusal contact area (139.5 ± 52.3 vs. 99.8 ± 43.8 μm; P = 0.043). Both design methods had similar mean and minimum occlusal clearances by the OR method (P > 0.05). For the evaluation in the master casts, both design techniques had similar mean and minimum occlusal clearances, number and distribution of occlusal contacts, and lateral interference ratios (P > 0.05).

CONCLUSION

Occlusal contact and clearance of posterior implant-supported single crowns designed by the CAD method can be at least as good as those designed by the conventional wax-up method.

Comparison of occlusal contact errors of 6 chairside CAD/CAM crowns: a self-controlled clinical study

OBJECTIVES

The purpose of this self-controlled clinical study was to evaluate the trueness of occlusal contacts of chairside CAD/CAM crowns fabricated using different ceramic materials.

MATERIALS AND METHODS

Ten volunteers (8 females, 2 males, average aged 20-30) were recruited in this study registered in the Clinical Trials Registry (#NCT05346744). After tooth preparation on tooth 36 or 46, an optical scan unit (CEREC Omnicam, Sirona Dental Systems, Germany) was utilized to perform the intraoral scanning. For each volunteer, 6 crowns were fabricated via the chairside CAD/CAM process (CEREC, Sirona Dental Systems, Germany) using the following materials: InCoris TZI (ZIR), Celtra Duo (CD), e.max CAD (EMA), UP.CAD (UP), Enamic (ENA) and Hyramic (HY). The microhardness of the milled surfaces was measured through a Vickers hardness Tester (HVS-50Z, Trojan, China). Together with the amount of occlusal adjustment, the occlusal contact trueness at both milling and postprocessing stages were quantitatively analyzed by using a high-precision scanner (ATOS, GOM Technologies, USA) and a reverse engineering software (Geomagic Control, 3D Systems, Rock Hill, SC). The times of chairside occlusal adjustment were also recorded. Data were analyzed by one-way analysis of variance (ANOVA) and ANOVA with randomized block design followed by Bonferroni test (p = 0.05).

RESULTS

Significant differences were found in surface microhardness of the materials tested (CD > EMA ~ UP > ENA > HY > ZIR, p < 0.05). Regarding the occlusal contact trueness at milling stage, ZIR showed the lowest (p < 0.05), while no significant difference was found among others (p > 0.05). At postprocessing stage, except for ENA, the occlusal contact trueness of ZIR was significantly lower than that of others (p < 0.05). As for occlusal adjustment amount, ZIR was lower than CD and ENA (p < 0.05). In addition, ENA, HY and ZIR required fewer times of occlusal adjustment than EMA, UP and CD.

CONCLUSIONS

The type of chairside CAD/CAM ceramic materials affected the occlusal contact trueness throughout the process of manufacturing and the amount of clinical occlusal adjustment as well. Zirconia exhibited the worst occlusal contact trueness at both milling and postprocessing stages, while the amount of occlusal adjustment was the least, which recommended the least occlusal compensation.

CLINICAL RELEVANCE

For better trueness, different negative occlusal offsets are suggested to be applied in the design process, so as to suit the material of the restoration.

Artificial intelligence-designed single molar dental prostheses: A protocol of prospective experimental study

BACKGROUND

Dental prostheses, which aim to replace missing teeth and to restore patients' appearance and oral functions, should be biomimetic and thus adopt the occlusal morphology and three-dimensional (3D) position of healthy natural teeth. Since the teeth of an individual subject are controlled by the same set of genes (genotype) and are exposed to mostly identical oral environment (phenotype), the occlusal morphology and 3D position of teeth of an individual patient are inter-related. It is hypothesized that artificial intelligence (AI) can automate the design of single-tooth dental prostheses after learning the features of the remaining dentition.

MATERIALS AND METHODS

This article describes the protocol of a prospective experimental study, which aims to train and to validate the AI system for design of single molar dental prostheses. Maxillary and mandibular dentate teeth models will be collected and digitized from at least 250 volunteers. The (original) digitized maxillary teeth models will be duplicated and processed by removal of right maxillary first molars (FDI tooth 16). Teeth models will be randomly divided into training and validation sets. At least 200 training sets of the original and the processed digitalized teeth models will be input into 3D Generative Adversarial Network (GAN) for training. Among the validation sets, tooth 16 will be generated by AI on 50 processed models and the morphology and 3D position of AI-generated tooth will be compared to that of the natural tooth in the original maxillary teeth model. The use of different GAN algorithms and the need of antagonist mandibular teeth model will be investigated. Results will be reported following the CONSORT-AI.

The occlusal precision of milled versus printed provisional crowns

OBJECTIVES

The aim of this study was to compare the occlusal precision of computer-aided-design/ computer-assisted-manufacturing (CAD/CAM) milled versus 3D printed polymethylmethacrylate (PMMA) temporary prosthetic crowns , starting from the same digital CAD design.

MATERIALS AND METHODS

The study sample included 34 patients presenting 34 premolars in need of prosthetic rehabilitation: a total of 68 temporary crowns were manufactured, 34 of which milled and 34 printed. Immediately after manufacturing, the milled and printed provisionals were scanned with a desktop scanner (E1, 3Shape) to obtain STL files, that were superimposed to the original CAD design in order to identify the occlusal trueness (Analysis A). A second occlusal comparison was performed by scanning both kind of provisional after being placed intraorally with Trios scanner (3 Shape); intraoral scans were obtained in order to compare STL files of provisionals before and after occlusal adjustments (Analysis B).  The occlusal trueness was identified at three reference points, P1 (vestibular cusp), P2 (palatal / lingual cusp), P3 (central fossa). The statistical analysis was performed using the R 3.4.3 statistical software (The R Foundation for Statistical Computing), with a significance level of p <0.05.

RESULTS

Overall, the printed crowns showed lower occlusal differences than the milled crowns, when compared to the CAD design file, with a statistically significant difference in P1 (difference of 0.025 ± 0.046 mm), P2 (difference of 0.027 ± 0 0.044 mm) and P3 ( difference of 0.018 ± 0.050) for Analysis A (p <0.05). In the Analysis B the direct comparison between the degree of average difference between the pre-adjustment and the post-occlusal adjustment of the milled and printed crowns shows that the printed crowns have lower occlusal mean difference values in all three points compared to the milled ones (difference of 0.146 ± 0.273 mm in P1, 0.285 ± 0.360 mm in P2 and 0.257 ± 0.277 mm in P3).

CONCLUSIONS

Within the limitation of this study, the data obtained showed a better occlusal surface dimensional accuracy of the 3D printed provisional crowns, when compared to the milled ones. Comparing the results obtained, it is possible to assume that the intraoral scans also had a contribution to occlusion, beside the manufacturing method. 3D printing can be successfully applied for manufacturing temporary PMMA crowns.

Effect of occlusal anatomy of CAD/CAM feldspathic posterior crowns in the stress concentration and fracture load

OBJECTIVES

This study evaluated the effect of restoration occlusal design on the maximum fracture load and stress distribution of a feldspathic ceramic crown.

MATERIALS AND METHODS

Twenty dentin analogues were used to simulate a full-crown preparation. Next, 20 feldspathic crowns were milled according to the occlusal design parameter available in the CAD database (Young or Adult). The crowns were cemented with dual cure resin-cement and loaded until fracture at 1 mm/min crosshead speed. Data were analyzed by using one-way ANOVA and Tukey tests (p < 0.05). The same geometry and experimental setup was modeled and exported to the computer aided engineering software and tensile stress concentration was calculated using the finite element method with 300 N occlusal load simulation.

RESULTS

The occlusal anatomy significantly influenced the load-to-fracture (p < 0.05). Adult design showed higher mean values (1149 ± 201 N) than Young design (454 ± 77 N). The maximum principal stress criteria showed similar stress pattern for both designs, however, the highest stress concentration was calculated for Young design (91 MPa) in the occlusal surface.

CONCLUSIONS

An anatomy design with reduced cusp angulation and less evident occlusal sulcus can reduce the stress concentration and increase the fracture load for feldspathic CAD/CAM posterior crowns.

Online dental teaching practices during the COVID-19 pandemic: a cross-sectional online survey from China

Background

Coronavirus disease 2019 (COVID-19) emerged in China in December 2019. The COVID-19 pandemic hindered dental education, as school buildings were closed. Online dental teaching provided an alternative teaching tool for dental education. However, the efficiency of online dental teaching and student preferences for online dental teaching are unclear.

Aim

To investigate the satisfaction with online dental teaching practices among undergraduate dental students and standardized resident physician training students during the COVID-19 pandemic in China.

Methods

A total of 104 undergraduate dental students and 57 standardized resident physician training students from Zhejiang University participated in the study. A 12-item survey was conducted. This investigation included the teaching methods received, frequency of classes, degree of satisfaction, preferred teaching method, whether to participate in a course regarding COVID-19 prevention, and the effects of teaching. The percentages were then calculated and evaluated for each item.

Results

A total of 161 students (104 undergraduate dental students and 57 standardized resident physician training students) participated in this survey. All students had online dental classes during the COVID-19 pandemic. Lecture-based learning (LBL), case-based learning (CBL), problem-based learning (PBL), team-based learning (TBL), and research-based learning (RBL) were selected as teaching methods. Students were more satisfied with LBL and CBL than PBL, RBL, and TBL. The majority of students had more than four classes per week. The most selected protective measures were hand washing, wearing masks, and wearing gloves. A total of 46.6% of students participated in courses on COVID-19. After training, the students consciously chose to wear face shields and protective clothing.

Conclusions

Dental students accepted online dental learning during the COVID-19 pandemic. Students preferred LBL and CBL and were satisfied with the classes. Courses on COVID-19 helped students understand how to prevent COVID-19 transmission in the dental clinic.

Design of occlusal wear facets of fixed dental prostheses driven by personalized mandibular movement

STATEMENT OF PROBLEM

Existing virtual articulators simulate mandibular movement by using various parameters and are used to design restorations. However, they are not able to reproduce actual patient movements, and the designs of occlusal wear facets by them and by personalized mandibular movement have not been compared.

PURPOSE

The purpose of this clinical study was to establish a clinical application protocol for a virtual articulator based on previous research and to evaluate the accuracy of the occlusal wear facets designed by it.

MATERIAL AND METHODS

The gypsum casts of 12 participants were scanned with a cast scanner as the original data. A single crown, 3-unit splinted crowns, a 5-unit fixed partial denture, and a fixed complete denture were virtually prepared on the digital mandibular casts by using the Geomagic Studio 2013 software program. High points were created at the wear facets, and corresponding digital wax patterns with occlusal interferences were generated. The exocad software program was used to design corresponding restorations with the copy method. Static (STA restoration) and dynamic (DYN restoration) occlusal adjustments were carried out with the built-in virtual articulator. The mandibular movements of participants were recorded by the novel virtual articulator system, and the occlusal surfaces of the digital wax patterns were adjusted (FUN restoration). The restorations adjusted with the 3 methods were compared with the original data. The mean value and root mean square (RMS) of 3D deviation and positive volumes (V+) in the occlusal direction were measured. Depending on the normality, 1-way ANOVA and the Kruskal-Wallis test were used to analyze the influence of occlusal surface design methods on the morphology of occlusal wear facets (α=.05).

RESULTS

The mean deviation of the 4 kinds of STA restorations ranged from 0.19 mm to 0.22 mm, the DYN restorations from 0.13 mm to 0.17 mm, and the FUN restorations from 0.03 mm to 0.09 mm. A significant difference was found between the STA and FUN restorations of the 3-unit splinted crowns and 5-unit fixed partial dentures (Ρ=.013, Ρ=.021). The mean values of 3D deviation and V+ decreased from the STA group to the DYN group and then to the FUN group. The RMS and V+ were statistically similar (Ρ>.05).

CONCLUSIONS

The preliminary results of the study indicate that the FUN 3-unit splinted crowns and 5-unit fixed partial dentures designed with the self-developed virtual articulator were better than the STA restorations. The FUN restorations were more coincident with the morphology of the wear facets on the original teeth.

Three-dimensional printed models versus conventional stone models

Aim: To compare the accuracy (trueness and precision)of cost-accessible three-dimensional (3D) printed models.Methods: A maxillary typodont (MM) was scanned andprinted 10 times in polylactic acid, resulting in 10 digitalmodels (DMs). Polyvinylsiloxane impressions were made toobtain 10 conventional stone models (SMs). All models werescanned and imported to CloudCompare software. The totalarea and three locations of interest were evaluated (zenith toincisal [Z-I], canine to canine [C-C], and first molar to canine[1M-C] distances). Total area evaluations were performed byaligning the MM and experimental models using the best-fitalgorithm and were compared using the Haussdorf distance.The distances between points of interest were measured usingthe point-picking tool at the same 3D coordinates. The meanvolumetric deviations were considered for trueness analysis.Precision was set as the standard deviation. Statisticaldifferences were evaluated using the Student’s t-test. Results:Total area volumetric comparisons showed that DMs showedsuperior trueness and precision (-0.02 ± 0.03) compared tothe SMs (0.37 ± 0.29) (P < 0.001). No differences between themodels were observed for Z-I (P = .155); however, SMs showedfewer deviations for C-C (P = .035) and 1M-C (P = .001) thanDMs. Conclusions: The DMs presented superior trueness andprecision for total area compared to the SMs; however, the SMswere more accurate when points of interest were evaluated.