Search

Accurately segmenting teeth and identifying the corresponding anatomical landmarks on dental mesh models are essential in computer-aided orthodontic treatment. Manually performing these two tasks is time-consuming, tedious, and, more importantly, highly dependent on orthodontists’ experiences due to the abnormality and large-scale variance of patients’ teeth. Some machine learning-based methods have been designed and applied in the orthodontic field to automatically segment dental meshes (e.g., intraoral scans). In contrast, the number of studies on tooth landmark localization is still limited. This paper proposes a two-stage framework based on mesh deep learning (called TS-MDL) for joint tooth labeling and landmark identification on raw intraoral scans. Our TS-MDL first adopts an end-to-end iMeshSegNet method (i.e., a variant of the existing MeshSegNet with both improved accuracy and efficiency) to label each tooth on the downsampled scan. Guided by the segmentation outputs, our TS-MDL further selects each tooth’s region of interest (ROI) on the original mesh to construct a light-weight variant of the pioneering PointNet (i.e., PointNet-Reg) for regressing the corresponding landmark heatmaps. Our TS-MDL was evaluated on a real-clinical dataset, showing promising segmentation and localization performance. Specifically, iMeshSegNet in the first stage of TS-MDL reached an averaged Dice similarity coefficient (DSC) at ${0.964}\pm {0.054}$ , significantly outperforming the original MeshSegNet. In the second stage, PointNet-Reg achieved a mean absolute error (MAE) of ${0}.{597}\pm {0}.{761} \, mm$ in distances between the prediction and ground truth for 66 landmarks, which is superior compared with other networks for landmark detection. All these results suggest the potential usage of our TS-MDL in orthodontics. [ABSTRACT FROM AUTHOR]

To compare the normal eruption pattern and angulation in impacted maxillary canines using panoramic radiographs to predict maxillary canine impaction. Patients aged 6 to 15 years were classified into the normal eruption group (n = 229) and the impaction group (n = 191). At least two panoramic radiographs were taken in the normal eruption group during the eruption process of the maxillary canine. The growth pattern of the maxillary canine was analyzed using an XY coordinate system, with the tip of the maxillary lateral incisor as the origin and the tooth's long axis as the Y-axis and measurement of the relative position of the crown tip and angulation of the maxillary canine. The crown tips of normally erupted maxillary canines were intensively distributed along the distal surface of the maxillary lateral incisor, while those of impacted canines were widely distributed. The angulations of the normally erupted canines increased as eruption increased along the lateral incisor and then decreased at the cervical point of the lateral incisor. The angulations of the impacted canines were scattered, with no uniform pattern. While using the normal eruption path of the maxillary canine and the pattern of change in angulation based on the distal surface of the maxillary lateral incisor, early intervention or regular follow-up is needed to prevent maxillary canine impaction. [ABSTRACT FROM AUTHOR]

Precisely labeling teeth on digitalized 3D dental surface models is the precondition for tooth position rearrangements in orthodontic treatment planning. However, it is a challenging task primarily due to the abnormal and varying appearance of patients’ teeth. The emerging utilization of intraoral scanners (IOSs) in clinics further increases the difficulty in automated tooth labeling, as the raw surfaces acquired by IOS are typically low-quality at gingival and deep intraoral regions. In recent years, some pioneering end-to-end methods (e.g., PointNet) have been proposed in the communities of computer vision and graphics to consume directly raw surface for 3D shape segmentation. Although these methods are potentially applicable to our task, most of them fail to capture fine-grained local geometric context that is critical to the identification of small teeth with varying shapes and appearances. In this paper, we propose an end-to-end deep-learning method, called MeshSegNet, for automated tooth labeling on raw dental surfaces. Using multiple raw surface attributes as inputs, MeshSegNet integrates a series of graph-constrained learning modules along its forward path to hierarchically extract multi-scale local contextual features. Then, a dense fusion strategy is applied to combine local-to-global geometric features for the learning of higher-level features for mesh cell annotation. The predictions produced by our MeshSegNet are further post-processed by a graph-cut refinement step for final segmentation. We evaluated MeshSegNet using a real-patient dataset consisting of raw maxillary surfaces acquired by 3D IOS. Experimental results, performed 5-fold cross-validation, demonstrate that MeshSegNet significantly outperforms state-of-the-art deep learning methods for 3D shape segmentation. [ABSTRACT FROM AUTHOR]

Objective: Visual differentiation of resin and tooth in a tooth cavity is not simple due to their highly similar shade. The purpose of the present study was to find any noninvasive method which can effectively differentiate resin from sound tooth in a resin-imbedded tooth for resin repair.Materials and methods: For the study, various resin products were imbedded into the cavity of sound tooth. By applying laser of different wavelengths, autofluorescence (AF) of sound tooth and resin products were obtained. Microhardness, X-ray radiograph, and DIAGNOdent were tested for each tooth, resin product, and resin-imbedded tooth.Results: For the AF spectra obtained using the 405-nm wavelength, sound tooth has emission peak at 440-470 nm and near 490 nm. Sound tooth has several times higher microhardness than resin products regardless of position in tooth subsurface. Due to the difference of radiopaque fillers' composition and concentration, resin products have different brightness in the X-ray radiograph. DIAGNOdent readings for tooth and resin products were inconsistently different, and the difference of obtained values was slightly not to be applicable for the differentiation.Conclusion: Among the tested methods, with noninvasive treatment, AF spectrum by the 405-nm wavelength showed the apparent difference between resin and tooth.Clinical significance: For the resin repair in a resin-imbedded tooth cavity, AF spectrum produced by 405-nm wavelength could be a useful method in tracing the resin-tooth boundary if combined with conventional X-ray radiography. [ABSTRACT FROM AUTHOR]

Abstract: Objective: To compare the cuspal strain in Class II restorations made with bulk‐fill and conventional composite resins. Materials and methods: Fifty extracted maxillary premolars were mounted into phenolic rings and divided into five groups (n = 10). Specimens received standardized MOD preparations. A two‐step self‐etch adhesive was applied and the preparations were restored using a custom matrix as follows: Filtek Supreme Ultra in eight 2‐mm increments (FSUI); Filtek Supreme Ultra in bulk (FSUB); SonicFill in bulk (SF); SureFil SDR flow in bulk, covered with a 2‐mm occlusal layer of Filtek Supreme Ultra (SDR/FSU); Tetric EvoCeram Bulk Fill in bulk (TEBF). Strain gages bonded to the buccal and lingual cusps recorded cuspal strain during restorations. End strain values were determined and data were subjected to Kruskal‐Wallis testing, followed by one‐way ANOVA and Tukey´s post hoc test. Results: Combined strain values and standard deviations (in µɛ) were: FSUI: 723 ± 102.8, FSUB: 929.2 ± 571.9, SF: 519.1 ± 80.2, SDR‐FSU: 497.4 ± 67.6 and TEBF: 604.5 ± 127.1. A significant difference was found between group FSUI and groups SF, SDR‐FSU, and TEBF. Group FSUB showed significantly higher mean strain and greater standard deviation than all other groups due to cuspal fractures, and was thus excluded from the statistical analysis. Conclusions: The tested bulk‐fill composite resins exerted less strain onto tooth structure than the incrementally placed conventional composite resin, although the magnitude of generated strain was product‐dependent. Bulk‐filling with conventional composite resins is contraindicated. Clinical significance: Bulk‐fill composite resins exerted less strain onto adjacent tooth structure than a traditional composite, even when that composite is was placed incrementally. Bulk‐filling with traditional composite resins is unpredictable and contraindicated. [ABSTRACT FROM AUTHOR]

The author discusses the result of a study which evaluates the vertical positions of anterior teeth on smile esthetics determined by the Japanese and Korean orthodontists and orthodontic patients. The author states that the anterior teeth's vertical positions were assessed after digital modification in 1-millimeter (mm) increments, from 5mm upper lip coverage to 5mm gingival exposure. He notes that orthodontists and the female patients shared the same preference in the smile evaluation.