[Dental X-ray diagnostics with the orthopantomography - Technique and typical imaging results]

STS-TransUNet: Semi-supervised Tooth Segmentation Transformer U-Net for dental panoramic image

In this paper, we introduce a novel deep learning method for dental panoramic image segmentation, which is crucial in oral medicine and orthodontics for accurate diagnosis and treatment planning. Traditional methods often fail to effectively combine global and local context, and struggle with unlabeled data, limiting performance in varied clinical settings. We address these issues with an advanced TransUNet architecture, enhancing feature retention and utilization by connecting the input and output layers directly. Our architecture further employs spatial and channel attention mechanisms in the decoder segments for targeted region focus, and deep supervision techniques to overcome the vanishing gradient problem for more efficient training. Additionally, our network includes a self-learning algorithm using unlabeled data, boosting generalization capabilities. Named the Semi-supervised Tooth Segmentation Transformer U-Net (STS-TransUNet), our method demonstrated superior performance on the MICCAI STS-2D dataset, proving its effectiveness and robustness in tooth segmentation tasks.

Enigmatic Formations Found in Routine Orthopantomography (OPG) Examinations: A Case Report

We describe two clinical cases of occasional radiographic findings on orthopantomography (OPG) that were performed routinely, for which the definitive diagnosis may be uncertain. After an accurate remote and recent anamnesis, for reasons of exclusion, we hypothesize a rare case of the retention of a contrast medium in the parenchyma of the major salivary glands (parotid, submandibular, and sublingual) and their excretory ducts as a consequence of sialography examination. In the first case we analyzed, we found it difficult to classify the radiographic signs on the sublingual glands, left parotid, and submandibular, while in the second case, only the right parotid was involved. Using CBCT, the spherical findings were highlighted, with multiple having different dimensions, as well as radiopaque in their peripheral portion and more radiolucent inside them. We could immediately exclude salivary calculi, which usually have a more elongated/ovoid shape and appear homogeneously radiopaque without radiolucency areas. These two cases (of hypothetic medium contrast retention with unusual and atypical clinical-radiographic presentation) have very rarely been comprehensively and correctly documented in the literature. No papers have a follow-up longer than 5 years. We conducted a review of the literature on the PubMed database, finding only six articles reporting similar cases. Most of them were old articles, demonstrating the low frequency of this phenomenon. The research was performed using the following keywords: "sialography", "contrast medium", "retention" (six papers) and "sialography", and "retention" (13 papers). Some articles were present in both searches, and the really significant ones (defined after a careful reading of the entire article and not only of the abstract) resulted only in six occurrences in a time span from 1976 to 2022.

Imaging furcation defects with low-dose cone beam computed tomography

Different cone beam computed tomography (CBCT) protocols have shown promising results for imaging furcation defects. This study evaluates the suitability of low-dose (LD)-CBCT for this purpose. Fifty-nine furcation defects of nine upper and 16 lower molars in six human cadavers were measured by a high-dose (HD)-CBCT protocol, a LD-CBCT protocol, and a surgical protocol. HD-CBCT and LD-CBCT measurements were made twice by two investigators and were compared with the intrasurgical measurements, which served as the reference. Furcation defect volumes generated from HD-CBCT and LD-CBCT imaging were segmented by one rater. Cohen's kappa and intraclass correlation coefficient (ICC) values were calculated to determine intra- and interrater reliability. The level of significance was set at α = 0.05. In total, 59 furcation defects of nine upper and 16 lower human molars were assessed. Comparing CBCT furcation defect measurements with surgical measurements revealed a Cohen's kappa of 0.5975 (HD-and LD-CBCT), indicating moderate agreement. All furcation defects identified by HD-CBCT were also detected by LD-CBCT by both raters, resulting in a Cohen's kappa of 1. For interrater agreement, linear furcation defect measurements showed an ICC of 0.992 for HD-CBCT and 0.987 for LD-CBCT. The intrarater agreement was 0.994(r1)/0.992(r2) for HD-CBCT and 0.987(r1)/0.991(r2) for LD-CBCT. The intermodality agreement was 0.988(r1)/0.991(r2). Paired t-test showed no significant differences between HD-CBCT and LD-CBCT measurements. LD-CBCT is a precise and reliable method for detecting and measuring furcation defects in mandibular and maxillary molars in this experimental setting. It has the potential to improve treatment planning and treatment monitoring with a far lower radiation dose than conventional HD-CBCT.

Low-dose CBCT imaging of alveolar buccal bone adjacent to mandibular anterior teeth- a pilot study

Objectives

Accurate description of buccal bone adjacent to mandibular anterior teeth is helpful for planning and monitoring periodontal and orthodontic treatment. Low-dose cone beam computed tomography (LD-CBCT) imaging has shown promising results for very small dental structures in animals. This study asserts that LD-CBCT is sufficiently accurate to measure buccal alveolar bone adjacent to human mandibular anterior teeth.

Materials and methods

Buccal bone level adjacent to 16 mandibular anterior teeth from four human cadavers was measured radiographically using one high-dose (HD) CBCT protocol and two LD-CBCT protocols. The resulting radiographic measurements of buccal bone height (bl) and thickness (bt) were compared with reference probe and reflected-light microscopy measurements. Measurement medians and Bland–Altman plots were calculated, and a linear mixed model was used to compare raters and imaging modalities.

Results

All regression coefficients were approximately 0, indicating high interrater, intrarater, and intermodality agreement. No significant differences were found between reference measurements and CBCT protocols. The mean differences for bl measurements were 0.07 mm (rater 1 [r1]) and 0.12 mm (rater 2 [r2]) for HD-CBCT; 0.07 mm (r1) and 0.13 mm (r2) for LD-CBCT-1; and 0.02 mm (r1) and 0.01 mm (r2) for LD-CBCT-2. For bt measurements, mean differences were 0.02 mm (r1) and 0.02 mm (r2) for HD-CBCT; 0.01 mm (r1) and 0.01 mm (r2) for LD-CBCT-1; and 0.00 mm (r1) and 0.01 mm (r2) for LD-CBCT-2.

Conclusions

Within the limitations of the present study, LD-CBCT seems to be a precise method for describing buccal bone and its thickness adjacent to mandibular anterior teeth in this experimental setting.

Clinical relevance

For the first time, this study showed LD-CBCT produces excellent results and is a reliable modality for imaging buccal bone in vitro. If clinical studies confirm these results, LD-CBCT could enable better treatment planning and monitoring at a radiation dose that is far lower than that of conventional HD-CBCT but similar to that of panoramic views.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00784-022-04389-x.

Ex-vivo imaging of buccal and oral periodontal bone with low-dose CBCT in porcine jaws

OBJECTIVES

New CBCT devices have been developed which can provide "low-dose CBCTs (LD-CBCT)". Aim of this study is to investigate the suitability of LD-CBCT for measurement of alveolar buccal/oral bone.

METHODS AND MATERIALS

Vestibular and oral bone loss of the teeth of seven porcine mandibles free of soft tissues were investigated by Micro-CT and three CBCT-modes: high-dose (HD), standard-dose (SD) and low-dose (LD). Radiographic measurements of bone loss (bl) and vestibular and oral bone thickness (bt) were made by two raters at 69 sites. Measurement means and differences, Intraclass correlation (ICC) and Bland-Altman plots were calculated.

RESULTS

ICCs between raters(r) concerning bl were 0.954 for HD, 0.949 for SD and 0.945 for LD; concerning bt they were 0.872 for HD, 0.845 for SD and 0.783 for LD. Means of differences of bt measurements were -0.01 mm(r1)/0.00 mm(r2) for HD, 0.04 mm(r1)/0.02 mm(r2) for SD and 0.02 mm(r1)/0.04 mm(r2) for LD; for bl measurements they were 0.06 mm(r1)/0.05 mm(r2) for HD, -0.01 mm(r1)/0.13 mm(r2) for SD and 0.07 mm(r1)/0.16 mm(r2) for LD.Linear regression indicates no noticeable differences between methods and the raters with respect to bl and bt.

CONCLUSIONS

Relating to the CBCT-device used in this study, LD-CBCT is a promising method to detect and describe buccal and oral periodontal bl and bt. Further studies with human anatomic structures must confirm these results.

FEATURES OF THE VARIATIVE MORPHOLOGY OF THE MANDIBULAR SECOND MOLAR USING CONE-BEAM COMPUTED TOMOGRAPHY IMAGING

OBJECTIVE

The aim: To characterize the variations in the anatomical structure of the second lower molar.

PATIENTS AND METHODS

Materials and methods: In order to determine the topography of the roots and root canals of the mandibular second molar, modern imaging methods were used: orthopan¬tomography and cone-beam computed tomography.

RESULTS

Case report: During the analysis of the orthopantomogram of the patient, the three-root morphology of the root system of 37 teeth was revealed. That is, three roots were clearly visualized. After detecting an atypical shape of the root system of the mandibular second molar in the panoramic image, the patient was offered to perform cone-beam computed tomography (CPCT) for diagnostic purposes. Analysis by cone-beam computed tomography showed the presence of four mature roots of normal length, which were arranged in the shape of a diamond.

CONCLUSION

Conclusions: The four-root configuration is considered a rare form for mandibular molars, but it should be considered in endodontic interventions and, if possible, in the planning of such interventions, use cone-beam computed tomography.