Dental caries is the most prevalent chronic condition worldwide. Early detection can significantly improve treatment outcomes and reduce the need for invasive procedures. Recently, near-infrared transillumination (TI) imaging has been shown to be effective for the detection of early stage lesions. In this work, we present a deep learning model for the automated detection and localization of dental lesions in TI images. Our method is based on a convolutional neural network (CNN) trained on a semantic segmentation task. We use various strategies to mitigate issues related to training data scarcity, class imbalance, and overfitting. With only 185 training samples, our model achieved an overall mean intersection-over-union (IOU) score of 72.7% on a 5-class segmentation task and specifically an IOU score of 49.5% and 49.0% for proximal and occlusal carious lesions, respectively. In addition, we constructed a simplified task, in which regions of interest were evaluated for the binary presence or absence of carious lesions. For this task, our model achieved an area under the receiver operating characteristic curve of 83.6% and 85.6% for occlusal and proximal lesions, respectively. Our work demonstrates that a deep learning approach for the analysis of dental images holds promise for increasing the speed and accuracy of caries detection, supporting the diagnoses of dental practitioners, and improving patient outcomes.

No abstract

The goals of the present study were to evaluate, in vitro, the staining of different composite resins submitted to different common beverages, and to compare the staining effect of each of these solutions. A total of 288 specimens were randomly divided into six groups and immersed for 4 weeks in five staining solutions represented by red wine, orange juice, coke, tea and coffee or in artificial saliva as a control group. When analyzed over a black background, mean ΔE00 values varied from 0.8 for Venus Diamond, Saremco Microhybrid and ELS in saliva and Estelite Posterior in coke to 37.6 for Filtek Supreme in red wine. When analyzed over a white background, mean ΔE00 values varied from 0.5 for Saremco Microhybrid in saliva to 51.1 for Filtek Supreme in red wine. All materials showed significant changes in color after 4 weeks of immersion in staining solutions. Significant differences were found between the tested composite resins and also between the staining solutions.

No abstract

The goal of this short communication is to present finite element analysis comparison of the stress distribution between CAD/CAM full crowns made of Lava Ultimate and of IPS e.max CAD, adhesively luted to natural teeth and to implant abutments with the shape of natural teeth. Six 3D models were prepared using a 3D content-creating software, based on a micro-CT scan of a human mandibular molar. The geometry of the full crown and of the abutment was the same for all models representing Lava Ultimate full crowns (L) and IPS e.max CAD full crowns (E) on three different abutments: prepared natural tooth (n), titanium abutment (t) and zirconia abutment (z). A static load of 400 N was applied on the vestibular and lingual cusps, and fixtures were applied to the base of the models. After running the static linear analysis, the post-processing data we analyzed. The stress values at the interface between the crown and the abutment of the Lt and Lz groups were significantly higher than the stress values at the same interface of all the other models. The high stress concentration in the adhesive at the interface between the crown and the abutment of the Lava Ultimate group on implants might be one of the factors contributing to the reported debondings of crowns.