BMC Oral Health最新文献

Background: Although caries is known to be influenced by diet and oral hygiene factors, susceptibility to the disease may be reduced by host factors, such as saliva antimicrobial peptides (AMPs). However, the relationship between salivary AMPs and early childhood caries (ECC) remains limited. Among the AMPs, cathelicidin LL-37 is mentioned in the literature for its bactericidal activity against oral pathogens, while statherin is recognized for its mechanism of reducing hydroxyapatite demineralization on the tooth surface. Cathelicidin LL-37 is a significant class of proteins within the immune system, and statherin prevents the precipitation of calcium phosphate in saliva, maintaining high levels of calcium for enamel remineralization and high phosphate levels for buffering.

Aim: This study aimed to compare the salivary levels of statherin, cathelicidin LL-37, and Streptococcus mutans (S. mutans), as well as the oral hygiene and dietary habits, of children with and without early childhood caries (ECC).

Design: A total of 24 children with ECC and 34 children without ECC were selected to participate in this study. The levels of S. mutans in unstimulated saliva collected from the children were measured by real-time polymerase chain reaction (PCR), and the levels of statherin and cathelicidin LL-37 in saliva were measured using an enzyme-linked immunosorbent assay (ELISA).

Results: There was no statistically significant difference in salivary statherin levels between the two groups, but salivary cathelicidin LL-37 levels were significantly higher in the ECC group (p = 0.024). The ECC group had significantly elevated S. mutans levels compared to the non-ECC group (p < 0.05). The frequency of sugar-containing foods and/or beverages consumed more than three times a day between meals was statistically significantly higher in the ECC group compared to the non-ECC group (p = 0.021). The statistical analysis also revealed a positive correlation between the dmfs score and age (p = 0.022).

Background: Since it has been widely acknowledged that oral care is effective in improving physical conditions, ward nurses are increasingly providing oral care for hospitalized patients in Japan. However, the current status of oral care practices and nurses' self-confidence in oral care skills has not been previously investigated. This study aimed to clarify these aspects among ward nurses.

Methods: A questionnaire survey was conducted in an acute care hospital, involving 18 ward departments and 644 nurses. The survey investigated the status of oral care implementation among the wards and nurses, and nurses' confidence levels regarding oral care skills. The impact of environmental factors within the wards on the implementation of oral care and nurses' confidence was analyzed through group comparisons. Additionally, multiple regression analysis was performed to identify factors associated with nurses' confidence.

Results: Responses were obtained from 16 wards and 189 nurses. Data from 160 nurses working in the 16 wards with environmental information were analyzed. Most nurses (91.9%) reported performing oral care daily. When wards were categorized into high- and low-need groups based on the proportion of patients requiring oral care, nurses in high-need wards were more likely to perform tongue cleaning, mouth wiping, oral mucosa care, moisturizing, and massage of perioral muscles. Furthermore, nurses working in the wards with an oral care manual and a designated core member for oral care reported higher confidence in their skills of oral care practices for patients with various oral problems, unconsciousness, and intubation. Additional multiple regression analysis of total self-confidence scores identified the presence of a core member, years of nursing experience, and experience of oral care training as factors positively associated with self-confidence.

Conclusions: Nurses in high-need wards reported providing more comprehensive oral care. The availability of oral care manuals, assignment of core members for oral care, and provision of oral care training were effective in enhancing nurses' confidence in their oral care skills. These findings suggest that, to further promote oral care practices by nurses, dental professionals are required to closely collaborate with nurses, especially core members, and provide guidance and training regarding oral care practices.

Objective: This study evaluated multiple deep learning approaches for automated classification of dental conditions in panoramic radiographs, comparing custom convolutional neural networks (CNNs), hybrid CNN-machine learning models, and fine-tuned pre-trained architectures, comparing the performance of custom convolutional neural networks (CNNs), hybrid CNN-machine learning models, and fine-tuned pre-trained architectures for detecting fillings, cavities, implants, and impacted teeth.

Methods: A dataset of 1,512 panoramic X-ray images with 11,137 manually annotated bounding boxes for four dental conditions (fillings, cavities, implants, and impacted teeth) was analyzed, with regions of interest extracted using expert annotations for subsequent AI-based classification. Class imbalance was addressed through random downsampling, creating a balanced dataset of 894 samples per condition. Multiple approaches were evaluated via 5-fold cross-validation: a custom CNN, hybrid models combining CNN features with traditional classifiers (Support Vector Machine, Decision Tree, Random Forest), and fine-tuned pre-trained networks (VGG16, Xception, ResNet50). Performance was assessed using accuracy, precision, recall, and F1-score metrics.

Results: The hybrid CNN-Random Forest model achieved the highest accuracy of 85.4 ± 2.3% with macro-F1 score of 0.843 ± 0.028, representing an 11% point improvement over the custom CNN (74.29% accuracy, 0.724 macro-F1). VGG16 demonstrated superior pre-trained architecture performance (82.3 ± 2.0% accuracy, 0.817 macro-F1), followed by Xception (80.9 ± 2.3%) and ResNet50 (79.5 ± 2.7%). CNN + Random Forest exhibited exceptional fillings detection (F1: 0.860 ± 0.033) with balanced multi-class performance. Systematic misclassifications between morphologically similar conditions revealed inherent diagnostic challenges.

Conclusion: Hybrid CNN-based approaches combining feature extraction with Random Forest classification provide superior discriminative capability for dental condition detection on manually annotated regions compared to standalone architectures. While computationally efficient hybrid models show promise as supportive diagnostic tools, observed misclassification patterns indicate these AI systems should serve as adjuncts to clinical expertise, requiring prospective validation studies.