Objectives: The present study aimed to train deep convolutional neural networks (CNNs) to detect caries lesions on Near-Infrared Light Transillumination (NILT) imagery obtained either in vitro or in vivo and to assess the models’ generalizability. Methods: In vitro, 226 extracted posterior permanent human teeth were mounted in a diagnostic model in a dummy head. Then, NILT images were generated (DIAGNOcam, KaVo, Biberach), and images were segmented tooth-wise. In vivo, 1319 teeth from 56 patients were obtained and segmented similarly. Proximal caries lesions were annotated pixel-wise by three experienced dentists, reviewed by a fourth dentist, and then transformed into binary labels. We trained ResNet classification models on both in vivo and in vitro datasets and used 10-fold cross-validation for estimating the performance and generalizability of the models. We used GradCAM to increase explainability. Results: The tooth-level prevalence of caries lesions was 41% in vitro and 49% in vivo, respectively. Models trained and tested on in vivo data performed significantly better (mean ± SD accuracy: 0.78 ± 0.04) than those trained and tested on in vitro data (accuracy: 0.64 ± 0.15; p < 0.05). When tested in vitro, the models trained in vivo showed significantly lower accuracy (0.70 ± 0.01; p < 0.01). Similarly, when tested in vivo, models trained in vitro showed significantly lower accuracy (0.61 ± 0.04; p < 0.05). In both cases, this was due to decreases in sensitivity (by −27% for models trained in vivo and −10% for models trained in vitro). Conclusions: Using in vitro setups for generating NILT imagery and training CNNs comes with low accuracy and generalizability. Clinical significance: Studies employing in vitro imagery for developing deep learning models should be critically appraised for their generalizability. Applicable deep learning models for assessing NILT imagery should be trained on in vivo data.

This narrative review aims to analyze sex-specific differences in oral health, examine hormonal influences in women during puberty, pregnancy, and menopause, and compare oral health behaviors between men and women. Articles were selected based on: (1) sex-specific aspects of oral diseases, (2) hormonal influences on oral health, (3) comparative analyses between sexes, and (4) sex-disaggregated data on oral disease prevalence. Women experience hormonal vulnerabilities with estrogen deficiency causing xerostomia, mucosal atrophy, and increased caries susceptibility, showing parallels between oral and vaginal mucosa. Men demonstrate higher periodontitis prevalence (57% vs. 38%), utilize preventive services one-third less frequently, and show higher smoking rates (67% vs. 42%) and traumatic dental injuries (2:1 ratio). Women maintain better oral hygiene and treatment adherence. Sex-specific factors affecting both sexes remain unconsidered in dental practice. Women need targeted interventions during hormonal transitions, while men require improved preventive care engagement. Future research integrating sex-specific considerations is required to enable personalized oral health approaches for both sexes.

After selective carious tissue removal, residual carious lesions remain radiographically detectable. Radiopaque tagging resolves the resulting diagnostic uncertainty but impedes bond strengths of adhesives to tagged dentin. We developed a protocol mitigating these detrimental effects. A 30%/50%/70% SnCl₂ solution was dissolved in distilled water or a 30%/50%/90% ethanol solution (E30/60/90) and applied to artificially induced dentin lesions. Tagging effects were radiographically evaluated using transversal wavelength-independent microradiography (n = 6/group). Groups with sufficient tagging effects at the lowest SnCl₂ concentrations were used to evaluate how tagging affected the microtensile bond strength of a universal adhesive (Scotchbond Universal) to sound and carious dentin (n = 10/group). Two different protocols for removing tagging material were tested: 15 s phosphoric acid etching and 5 s rotating brush application. Scanning/backscattered electron microscopy (SEM/BSE) and energy-dispersive X-ray spectroscopy (EDS) were used to assess surfaces after tagging and removal. The most promising removal protocol was revalidated microradiographically. Tagging significantly increased the radiopacity, with consistent effects for 30% SnCl₂ dissolved in water or E30. Microscopically, tagged surfaces showed a thick carpet of SnCl₂, and tagging reduced bond strengths significantly on carious dentin but not on sound dentin (p < 0.01). On carious dentin, removal of tagging material using acid etching and rotating brush was microscopically confirmed. Acid etching also mitigated any bond strength reduction (median: 21.3 MPa; interquartile range: 10.8 MPa) compared with nontagged dentin (median: 17.4 MPa; interquartile range: 20.6 MPa). This was not the case for brushing (median: 13.2 MPa; interquartile range: 13.9 MPa). Acid etching minimally reduced the radiographic tagging effect (p = 0.055). Phosphoric acid etching reduces the detrimental bond-strength effects of tagging without significantly decreasing radiographic tagging effects when using a universal adhesive.

Objectives: This systematic review analyzed the root canal morphology and configuration (RCC) of maxillary first premolars (Mx1Ps) and sex-specific differences based on existing literature. Methods: Registered in PROSPERO (CRD42023394460) and following PRISMA guidelines, systematic searches were conducted in five databases (Cochrane, Embase, LILACS, Scopus, MEDLINE via PubMed) using predefined MeSH terms. Additional studies were identified through cross-referencing. Studies on Mx1P RCCs were included, assessed using the AQUA tool. Data extraction focused on RCC prevalence, sex differences, root variations, and examination methods. Results: Of 865 studies, 86 were included, analyzing 31,325 teeth. The most common RCCs were 2-2-2/2 (IV, frequencies between 0.6–80.5%) and 1-1-1/1 (I, 1.1–72.0%). Mx1Ps primarily had two roots (7.1–96.2%) or one root (3.8–93.2%), with three-rooted variants being rare (0.4–6.5%). Males more frequently exhibited two- or three-rooted Mx1Ps with RCCs like 2-2-2/2 (IV; OR = 1.39 [1.22, 1.58]), and 1-1-3/3 (VIII; OR = 2.22 [1.59, 3.11]). Females showed higher frequencies of RCCs like 1-1-1/1 (I; OR = 0.71 [0.53, 0.96]), 2-2-1/1 (II; OR = 0.66 [0.57, 0.77]), 1-2-1/1 (III; OR = 0.70 [0.59, 0.83]), and 1-1-2/2 (V; OR = 0.81 [0.70, 0.95]). Conclusions: Mx1Ps predominantly have two roots and a 2-2-2/2 (IV) RCC. CBCT was the most used method, followed by staining and clearing. Clinicians should consider sex-specific and morphological variations.