International dental journal最新文献

Introduction and aims: The use of large language models (LLMs) in healthcare is expanding. Retrieval-augmented generation (RAG) addresses key LLM limitations by grounding responses in domain-specific, up-to-date information. This study evaluated RAG-augmented LLMs for infective endocarditis (IE) prophylaxis in dental procedures, comparing their performance with non-RAG models assessed in our previous publication using the same question set. A pilot study also explored the utility of an LLM as a clinical decision support tool.

Methods: An established IE prophylaxis question set from previous research was used to ensure comparability. Ten LLMs integrated with RAG were tested using MiniLM L6 v2 embeddings and FAISS to retrieve relevant content from the 2021 American Heart Association IE guideline. Models were evaluated across five independent runs, with and without a preprompt ('You are an experienced dentist'), a prompt-engineering technique used in previous research to improve LLMs accuracy. Three RAG-LLMs were compared to their native (non-RAG) counterparts benchmarked in the previous study. In the pilot study, 10 dental students (5 undergraduate, 5 postgraduate in oral and maxillofacial surgery) completed the questionnaire unaided, then again with assistance from the best performing LLM. Accuracy and task time were measured.

Results: DeepSeek Reasoner achieved the highest mean accuracy (83.6%) without preprompting, while Grok 3 beta reached 90.0% with preprompting. The lowest accuracy was observed for Claude 3.7 Sonnet, at 42.1% without preprompts and 47.1% with preprompts. Preprompting improved performance across all LLMs. RAG's impact on accuracy varied by model. Claude 3.7 Sonnet showed the highest response consistency without preprompting; with preprompting, Claude 3.5 Sonnet and DeepSeek Reasoner matched its performance. DeepSeek Reasoner also had the slowest response time. In the pilot study, LLM support slightly improved postgraduate accuracy, slightly reduced undergraduate accuracy, and significantly increased task time for both.

Conclusion: While RAG and prompting enhance LLM performance, real-world utility in education remains limited.

Clinical relevance: LLMs with RAG provide rapid and accessible support for clinical decision-making. Nonetheless, their outputs are not always accurate and may not fully reflect evolving medical and dental knowledge. It is crucial that clinicians and students approach these tools with digital literacy and caution, ensuring that professional judgment remains central.

Introduction and aims: This study investigates the feasibility of applying ChatGPT, a generative artificial intelligence (AI) language model, to develop a user-friendly carbon footprint calculator tailored for dental practices. Building on a previously developed Excel-based tool, the research aimed to evaluate ChatGPT's capacity to generate accurate emissions estimates and sustainability recommendations using different prompting strategies.

Methods: Three prompting variants were tested. Variant 1 employed an unstructured request to assess general responses. Variant 2 used structured data entry with predefined emission factors. Variant 3 combined structured input with instructions to rely exclusively on outputs from a previously validated sustainability tool. ChatGPT-generated results were compared with the Excel benchmark, focusing on accuracy, contextual relevance and alignment with peer-reviewed guidance.

Results: Unstructured prompts (Variant 1) produced general recommendations of limited contextual relevance. Structured prompts improved both accuracy and specificity. Variant 2 generated tailored outputs using emission factors, while Variant 3 provided detailed, evidence-based recommendations consistent with established literature. Across variants, ChatGPT's carbon footprint estimates were largely comparable to the Excel benchmark, with only minor discrepancies in waste-related emissions.

Conclusion: Structured prompting significantly enhances ChatGPT's performance in generating reliable carbon footprint data and recommendations for dental practices. When supported by transparent emission factors and credible literature, generative AI tools can increase access to environmental data, support sustainability decision-making and facilitate climate action in clinical contexts. However, limitations remain, including risks of inaccurate outputs ('hallucinations') and regional generalisations. Effective use requires prompt literacy and open access to validated emission factor databases to maximise impact and reliability.

Clinical relevance: AI-driven calculators such as ChatGPT can help dental teams without carbon accounting expertise to understand and reduce their environmental impacts, supporting the integration of sustainability into routine clinical practice.

Introduction: The Global Action Plan on Oral Health 2023-2030 is reaffirmed, promoting prevention, equitable access, and affordability of essential oral healthcare, aligned with universal health coverage and addressing social and commercial determinants of oral health. The plan aims for resilient health systems based on primary healthcare (PHC).

Objective: The objective of this study is to determine the frequency and distribution of the main oral pathologies treated in the establishments of the Ministry of Public Health of Ecuador between 2016 and 2022.

Methodology: This study employs a retrospective methodology, utilizing a database provided by the Ministry of Public Health of Ecuador of the RDACAA and PRAS applications, treated in the Qview program, and presented in Microsoft Excel 2019. The Ministry of Public Health of Ecuador uses the ICD-10 code for the coding of diagnoses, considering age, sex, ethnic self-identification, and priority groups.

Results: The results show that dentin caries (K02.1) is the most frequent pathology, followed by acute gingivitis (K05.0) and deposits on teeth (K03.6).

Conclusions: This study provides crucial information at a national level and proposes to be a pioneer in the planning and execution of oral health policies in Ecuador, suggesting a reformulation of the National Oral Health Plan.

Introduction and aims: Clinical fractures remain the primary cause of failure in dental all-ceramic restorations, highlighting the need to improve the mechanical performance and durability of ceramic material. This study aimed to develop a large language model (LLM)-based framework to automatically construct a structured database of dental ceramics and integrate it with machine learning (ML) to predict material properties and accelerate material design.

Methods: LLMs (Llama, Qwen, and DeepSeek) were employed to perform literature mining tasks, including text classification, information extraction from abstracts, and tabular data extraction. These processes were integrated into an automated pipeline to systematically extract and structure compositional and performance data from dental research articles. Ten ML algorithms were then trained using the curated database to establish predictive models of ceramic performance.

Results: In the classification task, a few-shot learning model with simple label prompts achieved an F1 score of 0.89. Fine-tuned LLMs achieved F1 scores exceeding 0.89 across various entity categories.ML models were developed to predict the classification of flexural strength, with the Extra Trees model performing best (F1 = 0.928), and external validation yielding F1 = 0.88. SHAP analysis identified ZrO₂ and SiO₂ as key contributor, and exhaustive search identified optimal compositional ranges.

Conclusions: This study demonstrates an AI-based pipeline combining LLM-driven data extraction and ML modelling, offering a scalable and accurate approach for accelerating the discovery and optimization of dental ceramics and other dental materials.

Clinical relevance: The findings underscore the potential of advanced LLMs and ML models in restorative dentistry and materials research.

Background: Periodontal disease is one of the most common oral diseases in the world and a significant public health challenge. Asia is the region with the highest number of cases. This study comprehensively analyzed the current situation and trends of periodontal diseases in Asia from 1990 to 2021, providing detailed insights into periodontal diseases in the region.

Method: The study employed the open data from the Global Burden of Disease 2021 Database to explore the characteristics of periodontal disease burden in Asia from 1990 to 2021, including the prevalence rate, incidence rate, and changes in disability-adjusted life years. A descriptive analysis of the burden of periodontal diseases in Asia was conducted from multiple dimensions, such as age, gender, and country. The autoregressive integrated moving average model was used to evaluate the trend from 2022 to 2036. Additionally, the Xtreme Gradient Boosting algorithm was used in conjunction with SHapley Additive Explanations for feature importance analysis and model interpretation.

Result: In 2021, there were approximately 684,742,467 patients with periodontal diseases in Asia. South Asia had the highest age-standardized prevalence rate and age-standardized disability-adjusted life years rate (ASDR), while Central Asia had the highest age-standardized incidence rate. The high-income Asia Pacific region exhibited the lowest age-standardized prevalence rate, age-standardized incidence rate, and ASDR. The age group with the highest number of patients was 50 to 54 years old, and the disease was more common in men. The Socio-Demographic Index was negatively correlated with periodontal diseases. The results of SHapley Additive Explanations analyses demonstrate that age is the most influential factor in predicting periodontal disease. Autoregressive integrated moving average model projections suggest that these indicators will remain stable through 2035, indicating that the overall burden of periodontal disease in Asia is expected to plateau rather than continue to rise.

Conclusion: The burden of periodontal diseases varies significantly across Asian regions. Therefore, future policy-making should fully consider differences among countries in disease epidemiological characteristics, medical resource distribution, and sociocultural backgrounds, and formulate targeted strategies to effectively reduce the burden of periodontal disease.

Background: Oral lichen planus (OLP), oral lichenoid lesions (OLL), and squamous cell carcinoma on a lichenoid background (SCC-over-LP/LLP) overlap clinically, delaying malignant transformation recognition.

Objective: To evaluate a multimodal large language model (ChatGPT-5) against oral medicine (OM) specialists for tripartite classification (OLP/OLL/SCC-over-LP/LLP) and malignant-risk flagging.

Methods: Cross-sectional, paired diagnostic accuracy study adhering to STARD/STARD-AI. Retrospective, anonymized cases (n = 262; OLP = 100, OLL = 100, SCC-over-LP/LLP = 62) were independently evaluated by ChatGPT-5 and a comparator panel of board-certified OM specialists using identical clinical histories and intraoral photographs (no histopathology provided to either). A separate reference standard panel (three OM experts) established the diagnosis using full clinical data and histopathology prior to index testing.

Primary outcome: paired accuracy (McNemar). Secondary: certainty (1-5), management agreement (Gwet's AC1), and recognition of malignant red-flag features.

Results: Overall accuracy was comparable (84.7% ChatGPT-5 vs 85.5% OM specialists; McNemar P = .856, Cohen's h = 0.03). Sensitivity was high for OLP 0.99 and SCC-over-LP/LLP 0.85; OLL sensitivity 0.70 with specificity 1.00. Biopsy/referral agreement was near-perfect (AC1 = 0.91). Malignant-risk features were correctly identified in 88% of SCC-over-LP/LLP cases by ChatGPT-5 vs 92% by OM specialists (P = .41).

Conclusions: A multimodal large language model can reach expert-level accuracy for OLP/OLL/SCC-over-LP/LLP and reliably flag malignant transformation risk, supporting its role as an adjunctive decision-support tool in OM.

Objectives: Bruxism is a masticatory muscle activity that can occur during sleep or wakefulness. Awake bruxism is often associated with psychosocial stress and conscious behaviours, while sleep bruxism is considered with a dominant central origin. Each entity has different clinical features, aetiologies, and therapeutic implications. The 7 October events in Israel marked a period of intense national distress, affecting large portions of the population. The aim of this study was to assess the prevalence of awake bruxism (AB) and sleep bruxism (SB) among individuals exposed to varying degrees of effect.

Methods: A total of 511 participants completed an online survey between December 2023 and April 2024. The survey included the following questionnaires: BruxScreen-Q (to assess AB and SB), Patient Health Questionnaire-4 (PHQ-4), and Brief Resilient Coping Scale. Participants were grouped according to displacement status, proximity to the conflict zone, and degree of personal impact. Statistical analyses, including chi-squared tests and logistic regression, were performed to identify associations between bruxism and psychological factors.

Results: The prevalence of AB was significantly higher among females (74.3%, P < .001), displaced respondents (52.5%, P = .046), and individuals present in the Gaza-border communities during the attacks (59.6%, P = .006). Regression analyses confirmed that anxiety (P = .001) and depression (P < .001) were significant predictors of AB. SB prevalence was higher among nondisplaced individuals (43.8% P < .032).

Conclusion: Phycological variables (anxiety and depression), and moderating situational factors, have a significant role in predicting AB. The findings underscore the need for further research on long-term psychological and physiological effects.

Clinical significance: This study reinforces the link between psychological stress and AB. Dental professionals should consider recent trauma and emotional distress as potential contributing factors when diagnosing and managing bruxism.