Explore the vast range of titles available.

Background Oral cancers can be preceded by clinically evident oral potentially malignant disorders (OPMDs). The current study evaluated the rate and the time of malignant transformation in the various OPMDs in a cohort of patients from southern Taiwan. Parameters possibly indicative for malignant transformation of OPMDs, such as epidemiological and etiological factors, and clinical and histopathological features were also described. Methods We followed-up 5071 patients with OPMDs—epithelial dysplasia with oral submucous fibrosis, epithelial dysplasia with hyperkeratosis/epithelial hyperplasia, hyperkeratosis/epithelial hyperplasia, oral submucous fibrosis, lichen planus, and verrucous hyperplasia—between 2001 and 2010 for malignant transformation. Results Two hundred nineteen of these 5071 OPMD patients (202 men, 17 women; mean age: 51.25 years; range: 30–81 years) developed oral cancers (179 squamous cell carcinomas; 40 verrucous carcinomas) in the same sites as the initial lesions at least 6 months after their initial biopsies. The overall transformation rate was 4.32% (mean duration of transformation: 33.56 months; range: 6–67 months). Additionally, the mean time of malignant transformation was significantly shorter for lesions with than without epithelial dysplasia. The risk of malignant transformation was 1.89 times higher for epithelially dysplastic than non-dysplastic lesions. The anatomical site of OPMD and the presence of epithelial dysplasia were significantly associated with malignant transformation. The hazard rate ratio was 1.87 times larger for tongue lesions than for buccal lesions. Conclusion Patients with OPMDs require long-term follow up.

Objectives Oral potentially malignant disorders (OPMDs) refer to a group of significant precursor lesions for oral squamous cell carcinoma (OSCC). Early differential diagnosis of its subtypes including oral leukoplakia (OLK), oral lichen planus (OLP) and oral submucous fibrosis (OSF) is crucial for timely prevention and management of OSCC. The diverse clinical features of aforementioned conditions render clinical diagnosis challenging. To facilitate accurate and efficient diagnosis, this study aimed to construct a digital pathology-based artificial intelligence (AI) platform based on a sample of 1080 cases. Materials and methods Four deep learning models–Twins-SVT, ResNet18, ResNet50, and Inception_v3–were employed for patch-level analysis. The AI platform was developed by fusing deep learning-derived predictive features with multiple machine learning algorithms. Results The AI platform developed in this study exhibited good diagnostic performance, with the diagnostic area under the receiver operating characteristic curve of 0.870 (95% CI: 0.827–0.913), 0.810 (95% CI: 0.751–0.870), and 0.833 (95% CI: 0.770–0.897) for OLK, OLP, and OSF, respectively. In addition, this study investigated the correlation between deep learning-derived AI features and pathological findings, demonstrating excellent interpretability of AI in pathology. Conclusions In this study, a robust AI platform was developed for differential diagnosis of OLK, OLP and OSF. Our research supports the feasibility of using deep learning methods to classify common OPMDs based on whole slide images.

AbstractThe many diverse terms used to describe the wide spectrum of changes seen in proliferative verrucous leukoplakia (PVL) have resulted in disparate clinical management. The objective of this study was to produce an expert consensus guideline for standardized assessment and reporting by pathologists diagnosing PVL related lesions. 299 biopsies from 84 PVL patients from six institutions were selected from patients who had multifocal oral leukoplakic lesions identified over several years (a minimum follow-up period of 36 months). The lesions demonstrated the spectrum of histologic features described in PVL, and in some cases, patients developed oral cavity squamous cell carcinoma (SCC). An expert working group of oral and maxillofacial and head and neck pathologists reviewed microscopic features in a rigorous fashion, in combination with review of clinical photographs when available. The working group then selected 43 single slide biopsy cases for whole slide digital imaging (WSI) review by members of the consensus conference. The digital images were then reviewed in two surveys separated by a washout period of at least 90 days. Five non-PVL histologic mimics were included as controls. Cases were re-evaluated during a consensus conference with 19 members reporting on the cases. The best inter-observer diagnostic agreement relative to PVL lesions were classified as “corrugated ortho(para)hyperkeratotic lesion, not reactive” and “SCC” (chi-square p = 0.015). There was less than moderate agreement (kappa < 0.60) for lesions in the “Bulky hyperkeratotic epithelial proliferation, not reactive” category. There was ≥ moderate agreement (> 0.41 kappa) for 35 of 48 cases. This expert consensus guideline has been developed with support and endorsement from the leadership of the American Academy of Oral and Maxillofacial Pathology and the North American Society of Head and Neck Pathologists to recommend the use of standardized histopathologic criteria and descriptive terminology to indicate three categories of lesions within PVL: (1) “corrugated ortho(para)hyperkeratotic lesion, not reactive;” (2) “bulky hyperkeratotic epithelial proliferation, not reactive;” and (3) “suspicious for,” or “squamous cell carcinoma.” Classification of PVL lesions based on a combination of clinical findings and these histologic descriptive categories is encouraged in order to standardize reporting, aid in future research and potentially guide clinical management.

Background Artificial intelligence (AI) has rapidly advanced in healthcare and dental education, significantly impacting diagnostic processes, treatment planning, and academic training. The aim of this study is to evaluate the performance differences between different large language models (LLMs) by analyzing their accuracy rates in answers to multiple choice oral pathology questions. Methods This study evaluates the performance of eight LLMs (Gemini 1.5, Gemini 2, ChatGPT 4o, ChatGPT 4, ChatGPT o1, Copilot, Claude 3.5, Deepseek) in answering multiple-choice oral pathology questions from the Turkish Dental Specialization Examination (DUS). A total of 100 questions from 2012 to 2021 were analyzed. Questions were classified as “case-based” or “knowledge-based”. The responses were classified as “correct” or “incorrect” based on official answer keys. To prevent learning biases, no follow-up questions or feedback were provided after the LLMs’ responses. Results Significant performance differences were observed among the models ( p  < 0.001). ChatGPT o1 achieved the highest accuracy (96 correct, 4 incorrect), followed by Claude (84 correct), Gemini 2 and Deepseek (82 correct each). Copilot had the lowest performance (61 correct). Case-based questions showed notable performance variations ( p  = 0.034), where ChatGPT o1 and Claude excelled. For knowledge-based questions, ChatGPT o1 and Deepseek demonstrated the highest accuracy ( p  < 0.001). Post-hoc analysis revealed that ChatGPT o1 performed significantly better than most other models across both case-based and knowledge-based questions ( p  < 0.0031). Conclusion LLMs demonstrated variable proficiency in oral pathology questions, with ChatGPT o1 showing higher accuracy. LLMs shows promise as a supplementary educational tool, though further validation is required.

In oral pathology, virtual clinical scenario illustrating dentist-patient interactions can be utilized by both students and health professionals to deliver/gain knowledge and make clinical diagnosis of oral lesions. Role play is also an educational technique which is designed to engage and motivate students in classrooms. This study aimed to compare usefulness of virtual clinical scenario and role play in learning oral pathology among second-year dental students. The students were randomly divided to one of the two groups: virtual clinical scenario group (n = 50) and role play group (n = 50). Virtual clinical scenario group was provided with virtual clinical cases of oral lesions through Google Forms whereas role play group was exposed to virtual clinical cases of oral lesions through role playing activity. Both groups underwent assessments before and after the intervention. Students’ perceptions on usefulness of both techniques in terms of diagnosis, visual parameters and impact on learning were evaluated by feedback questionnaire. Data were analyzed using Statistical Package for the Social Sciences version 27.0. Wilcoxon signed-rank test was used to compare pre-test and post-test scores. Additionally, the scores and students’ responses from both groups were compared using the Mann-Whitney U test. A P-value of < 0.05 was set as statistically significant. Students in both groups showed significantly higher post-test scores compared to their pre-test scores (P < 0.001). However, the role play group outperformed the virtual clinical scenario group, with a significantly higher post-test score (P = 0.04). Furthermore, feedback concerning role play was significantly higher than that for the virtual clinical scenario across multiple aspects (P < 0.05). Our findings suggest that role play emerges as the preferred method, significantly enhancing dental students’ learning experiences in oral pathology.

Oral Potentially Malignant Disorders (OPMDs), including Leukoplakia and Erythroplakia, carry significant risks of malignant transformation. Early differentiation from confounding inflammatory conditions like Oral Lichen Planus (OLP) and Candidiasis is critical yet challenging due to visual similarities. This study develops a robust, interpretable deep learning framework for automated multi-class classification of pre-localized oral lesions. A curated dataset of 705 high-resolution images across five categories (Leukoplakia, Erythroplakia, OLP, Candidiasis, and Normal) was utilized. We proposed a Multi-Architecture Weighted Ensemble Framework integrating ResNet-50, Xception, and EfficientNet-B0. A stratified patient-level splitting strategy prevented data leakage. Model interpretability and clinical utility were assessed via Gradient-weighted Class Activation Mapping (Grad-CAM) and Decision Curve Analysis (DCA). The ensemble model achieved 91.2% accuracy and a 90.8% macro-averaged F1-score, significantly outperforming individual baselines. The strategy improved OLP detection (F1-score: 0.83), effectively distinguishing it from Leukoplakia. Grad-CAM confirmed the model focuses on pathognomonic lesion features rather than confounding artifacts. DCA suggested a potential theoretical net clinical benefit over default strategies. This weighted ensemble framework demonstrates high retrospective accuracy and provides transparent visual explanations for the classification of pre-localized oral lesions. However, it must be interpreted strictly as a preliminary proof-of-concept investigation. While the current results suggest potential adjunctive value, extensive external validation, prospective testing, and clinician-in-the-loop studies are strictly necessary to validate its true clinical utility and impact on patient outcomes in primary care settings.

Background The role of microbes in diseases, especially cancer, has garnered significant attention. However, research on the oral microbiota in oral potentially malignant disorders (OPMDs) remains limited. Our study investigates microbial communities in OPMDs. Materials and methods Oral biopsies from19 oral leukoplakia (OLK) patients, 19 proliferative verrucous leukoplakia (PVL) patients, 19 oral lichen planus (OLP) patients, and 19 oral lichenoid lesions (OLL) patients were obtained. 15 SCC specimens were also collected from PVL patients. Healthy individuals served as controls, and DNA was extracted from their paraffin-embedded tissues. 2bRAD-M sequencing generated taxonomic profiles. Alpha and beta diversity analyses, along with Linear Discriminant Analysis effect size analysis, were conducted. Results Our results showed the microbial richness and diversity were significantly different among groups, with PVL-SCC resembling controls, while OLK exhibited the highest richness. Each disease group displayed unique microbial compositions, with distinct dominant bacterial species. Noteworthy alterations during PVL-SCC progression included a decline in Fusobacterium periodonticum and an elevation in Prevotella oris . Conclusions Different disease groups exhibited distinct dominant bacterial species and microbial compositions. These findings offer promise in elucidating the underlying mechanisms of this disease.

Oral potentially malignant disorders (OPMDs) comprise a group of oral mucosal abnormalities associated with an increased risk of malignant transformation. Early diagnosis, appropriate management, and long-term follow-up are essential to reduce the likelihood of cancer progression. This study aimed to evaluate the clinical and histopathological characteristics of OPMDs, their dysplastic changes, and clinical outcomes of patients referred to Alborz dental school between 2019 and 2025.This combined retrospective and prospective cohort study included 50 patients diagnosed with OPMDs. Data were collected using a standardized clinician-completed checklist, along with comprehensive clinical examinations and histopathological assessments. Recorded variables included demographic characteristics, type and location of lesions, and associated risk factors. The study population consisted of 64% females, with the majority of patients aged between 40 and 60 years. Plaque-type oral lichen planus and non-homogeneous leukoplakia were the most frequently diagnosed lesions. The buccal mucosa and tongue were the most commonly involved sites. Burning sensation was reported by 52% of patients. Histopathological examination revealed lichenoid mucositis in 86% of patients, with mild epithelial dysplasia observed in 7%. In contrast, dysplastic changes were identified in 57% of patients with leukoplakia. Therapeutic interventions included topical corticosteroids and intralesional corticosteroid injections. At follow-up, 80% of patients exhibited persistent lesions with plaque-type morphology and white lesions being most common; however, a subset of patients reported reduction in pain and burning sensation. Early diagnosis and structured long-term follow-up are essential for appropriate risk assessment and early detection of potential malignant progression

Artificial intelligence (AI) integration in diagnostic medicine has advanced accuracy and efficiency, particularly in pathology. This study assessed the diagnostic performance of three large language models (LLMs)—ChatGPT (GPT-4-turbo), Grok (xAI), and MANUS—in interpreting histopathology slides of oral lesions. A comparative diagnostic study was conducted using 100 high-resolution slides representing diverse oral pathologies. Images were sourced from a validated textbook and reviewed by two board-certified oral pathologists who provided consensus diagnoses. Each slide was analysed twice by the three AI models using standardized prompts. Diagnostic accuracy, intra-model consistency, inter-model concordance, and agreement with human experts were evaluated using descriptive statistics, Cohen’s kappa, McNemar’s test, and chi-square analysis. All AI models demonstrated high diagnostic accuracy. In the second round, Grok achieved the highest accuracy (97%), followed by MANUS (96%) and ChatGPT (94%). ChatGPT showed the highest intra-model consistency (κ = 0.918), while MANUS and Grok displayed substantial agreement (κ = 0.790 and 0.740). Expert pathologists achieved 98% accuracy. Comparisons between AI models and human diagnoses showed moderate to substantial agreement, with MANUS most aligned with experts. Most misclassifications occurred in histologically ambiguous cases, with no significant differences between AI models. Multimodal LLMs demonstrated strong diagnostic capabilities, consistency, and alignment with expert reasoning in oral histopathology interpretation. Grok was the most accurate, ChatGPT the most consistent, and MANUS the most expert-aligned. These findings support AI integration into digital pathology for diagnostic support, education, and quality assurance, with further validation in clinical datasets recommended.

Background Clinicopathological diagnosis and follow-up of oral lichen planus and leukoplakia are necessary due to its potential for malignant transformation and the need to differentiate it from other lichenoid diseases and proliferative verrucous leukoplakia. This study aimed to classify and compare sociodemographic and clinicopathological features among patients with oral lichen planus, oral lichenoid lesions and proliferative verrucous leukoplakia. Methods A transversal observational study in which oral leukoplakia and oral lichen planus patients were surveyed at the Oral Pathological Anatomy Service and Applied Biotechnology Laboratory was conducted. Sociodemographic and clinicopathological data were compared for the lesions studied with the chi-square test or Fisher’s exact test. Results After classification, 21 oral lichen planus lesions, 34 oral lichenoid lesions and 12 proliferative verrucous leukoplakia lesions were evaluated. Reticular patterns are more characteristic of oral lichen planus and plaque lesions of proliferative verrucous leukoplakia. The buccal mucosa was the most affected site in oral lichen planus lesions, and it was bilateral in all patients. Epithelial dysplasia was present in almost all patients with proliferative verrucous leukoplakia. Conclusions Compared with oral lichen planus and proliferative verrucous leukoplakia, oral lichenoid lesions presented intermediate features. This may delay proliferative verrucous leukoplakia diagnosis.