Search Results (59)

Inflammatory bowel disease (IBD) is associated with microbial dysbiosis, yet subtype-specific alterations in the tongue-coating microbiome remain insufficiently characterized. In this cross-sectional study, tongue-coating samples from 158 participants (94 healthy controls [HC], 19 ulcerative colitis [UC] patients, and 45 Crohn’s disease [CD] patients) were analyzed by 16S rRNA gene amplicon sequencing. We compared alpha and beta diversity, taxonomic composition, differential taxa, exploratory random-forest feature rankings, and SPIEC-EASI co-occurrence networks. Species richness did not differ significantly among groups, whereas Shannon and Simpson indices were lower in UC than in HC and CD. Bray–Curtis and Jaccard ordinations showed significant but partially overlapping community differences among the three groups. UC was characterized by enrichment of Proteobacteria, Neisseria, and Porphyromonass (p < 0.001), whereas CD showed higher relative abundances of Prevotella, Veillonella, Leptotrichia, and TM7x. Random forest and LEfSe analyses yielded concordant candidate discriminative taxa, but no independent validation cohort was available. Network analysis suggested group-specific co-occurrence patterns, with results interpreted as statistical associations rather than direct microbial interactions. These findings support the presence of subtype-associated tongue-coating dysbiosis in IBD and identify candidate taxa for future validation. Full article

Background: Tongue diagnosis is a fundamental diagnostic method in traditional medicine. Studies restricted to single ethnic groups may introduce bias and limit the clinical applicability of digital tongue diagnosis across diverse populations. Objectives: This study examined differences in tongue image features between Japanese and Brazilian (Caucasian ancestry) participants using digital tongue diagnosis technology and explored potential influencing factors. Methods: Tongue images were collected from 143 Japanese and 116 Brazilian participants attending traditional medicine clinics in Japan and Brazil. An independently developed tongue image analysis system (TIAS) was employed to extract shape, texture (gray level co-occurrence matrix), color (L*a*b color space), and deep-learning derived features (crack, prickle, tooth-mark, peel, greasy coating, stasis). Statistical analyses and machine learning models with SHAP explainability were used to compare features and identify key classification parameters. Results: Significant inter-group differences were observed in tongue shape, texture parameters (particularly at the root and tip), color parameters (especially middle-a-mean, middle-b-mean, tip-a-mean, and tip-b-mean), and deep features. The Japanese group showed a markedly higher prevalence of greasy coating (72.03% vs. 41.38%, p < 0.001) and stasis. Machine learning analysis revealed that the b value in the middle region of the tongue (middle-b-mean) contributed most strongly to the classification of greasy coating. Conclusions: The digital tongue image analysis system enables accurate and objective quantification of tongue features. Pronounced ethnic differences exist, particularly in the distribution of greasy coating. The middle-b-mean has the greatest impact on greasy coating classification. These findings underscore the importance of considering ethnic background when developing digital tongue diagnosis systems. Full article

Background/Objectives: Halitosis in geriatric patients is multifactorial, but the joint contribution of prosthetic rehabilitation type and polypharmacy after routine dental procedures has rarely been quantified. We investigated how prosthesis type, polypharmacy, and salivary function were associated with volatile sulfur compound (VSC) burden and self-perceived halitosis in elderly dental patients. Methods: This cross-sectional study enrolled 88 patients aged ≥65 years, four weeks after completing routine dental procedures. Participants were stratified into three groups: complete denture wearers (n = 30), partial removable denture wearers (n = 28), and fixed prostheses/implants (n = 30). We measured unstimulated salivary flow rate (uSFR), tongue coating index (TCI), denture biofilm index, total VSCs (Halimeter^®), organoleptic score (0–5), and self-perceived halitosis. Polypharmacy, comorbidities, and the Geriatric Oral Health Assessment Index (GOHAI) were recorded. Analyses included one- and two-way ANOVA, Spearman correlations, theory-informed multivariable linear and logistic regression, exploratory mediation analysis, and ROC curves. Results: Forty-two participants (47.7%) reported halitosis. Mean VSC differed across groups (complete dentures 278.2 ± 38.6 ppb; partial 211.2 ± 46.3 ppb; fixed 164.4 ± 43.9 ppb; ANOVA p < 0.001). uSFR correlated inversely with VSC (ρ = −0.61, p < 0.001) and TCI correlated positively (ρ = 0.56, p < 0.001). A significant prosthesis × polypharmacy interaction was observed (F = 3.74, p = 0.029, η²_p = 0.082): polypharmacy was associated with higher VSC most clearly among partial and fixed prostheses wearers, whereas complete denture wearers showed high VSC levels regardless of polypharmacy status. Exploratory mediation findings were consistent with partial indirect association, with 45.9% of the polypharmacy–VSC association statistically explained by reduced uSFR; however, the cross-sectional design precludes causal or temporal interpretation. The full multivariable model showed apparent discrimination for self-perceived halitosis (AUC = 0.92), while the simplified four-item chairside composite model showed AUC = 0.89; neither estimate was optimism-corrected or externally validated. Conclusions: In elderly post-procedure patients, complete denture wearing, polypharmacy, and salivary hypofunction were independently and jointly associated with higher halitosis burden. Reduced salivary flow was consistent with a partial indirect statistical pathway in the polypharmacy–VSC association, supporting hydration counseling and meticulous prosthesis hygiene as low-cost geriatric interventions. Sensitivity analyses excluding implant-supported restorations, participants with MMSE scores of 24–26, and expanded mediation models including TCI and biofilm/plaque did not materially change the main inference. Full article

Background and Objectives: Halitosis is a prevalent oral concern that meaningfully affects oral health-related quality of life (OHRQoL), yet how active dental treatment is associated with short-term changes in the objective–subjective halitosis–QoL nexus remains poorly quantified. Interpretation is complicated by the multifactorial nature of malodor and by baseline differences between patients selected for different dental procedures. We compared changes in volatile sulfur compound (VSC) emissions, organoleptic ratings, tongue-coating burden, and OHIP-14 across three contrasting treatment modalities and explored whether VSC change statistically accounted for OHRQoL change. Methods: In a non-randomized prospective comparative study, 119 adults (18–67 y) commencing one of three procedures were assessed at baseline and at 8 weeks: scaling and root planing (Group A, n = 42), fixed prosthodontic rehabilitation (Group B, n = 38), or fixed orthodontic appliance bonding (Group C, n = 39). Outcomes included Halimeter^® VSC (ppb), Rosenberg organoleptic score (0–5), Winkel tongue-coating index (TCI), self-perceived halitosis, and OHIP-14 total and seven-domain scores. Mixed-design ANOVA, ANCOVA, prespecified multivariable regression, mediation (5000 bootstrap resamples), receiver operating characteristic analysis, and four-class latent class analysis were performed. A sensitivity-analysis framework including expanded covariate adjustment, propensity-score overlap weighting, and baseline-severity strata was also applied to address residual baseline imbalance. Secondary mediation, ROC, and latent-class analyses were considered exploratory. Results: At 8 weeks, VSCs fell by 116.4 ± 38.7 ppb in Group A and 35.4 ± 29.1 ppb in Group B but rose by 34.3 ± 28.6 ppb in Group C (p < 0.001). OHIP-14 improved by 10.3 and 4.9 points in A and B and worsened by 3.7 in C (p < 0.001). ΔVSC correlated with ΔOHIP-14 (ρ = 0.51, p < 0.001) and most strongly with the psychological discomfort domain (ρ = 0.58). VSC change mediated 35.1% of the periodontal-versus-orthodontic association on QoL (indirect β = −4.7; 95% CI −6.3 to −3.1). Because VSC and OHIP-14 changes were measured over the same interval, mediation was interpreted cautiously. A ΔVSC threshold of −63 ppb predicted clinically meaningful OHIP-14 improvement (AUC = 0.81). Latent class analysis identified four distinct responder phenotypes. The cutoff and responder classes were internally derived and require external validation. Sensitivity analyses preserved the direction of the primary contrasts, but residual confounding remains possible. Conclusions: Treatment modality was associated with the direction and magnitude of halitosis and QoL change, with orthodontic patients constituting a vulnerable subgroup. Targeted oral-hygiene reinforcement during fixed-appliance therapy is warranted. Full article

Background and Objectives: Halitosis is a common condition with substantial psychosocial impact, frequently driven by intra-oral biofilm, tongue coating, and reduced salivary clearance. This study compared the short-term effectiveness of standardized counseling alone, probiotic lozenges containing Streptococcus salivarius K12, and a zinc-containing mouthrinse in adults with persistent intra-oral halitosis. Materials and Methods: In this 4-week, parallel-group, randomized pragmatic trial, 117 adults with bothersome halitosis for at least 3 months and baseline organoleptic score ≥ 2 were allocated 1:1:1 to standard care, probiotic lozenges, or zinc mouthrinse. All participants received standardized counseling and tongue cleaning instructions. The primary endpoint was change in volatile sulfur compounds (VSCs) measured by portable sulfide monitoring. Secondary outcomes included organoleptic score, Halitosis Associated Life-Quality Test (HALT), Oral Health Impact Profile-14 (OHIP-14), tongue coating, plaque, and salivary Solobacterium moorei quantified by qPCR. Results: Baseline demographic, clinical, and biochemical characteristics were comparable across groups. All interventions improved outcomes over 4 weeks, but improvements followed a consistent gradient favoring zinc mouthrinse, followed by probiotic lozenges, then standard care. Mean VSC reduction was −12.7 ± 33.9 ppb with standard care, −47.3 ± 42.2 ppb with probiotics, and −78.5 ± 36.3 ppb with zinc mouthrinse (p < 0.001). Organoleptic scores improved by −0.2 ± 0.7, −0.8 ± 0.8, and −1.2 ± 0.8, respectively (p < 0.001). HALT and OHIP-14 scores showed parallel reductions, and moderate/severe halitosis at week 4 remained most frequent in the standard care group (58.9%) and least frequent in the zinc group (20.5%; p = 0.004). Conclusions: Both active adjunctive strategies improved intra-oral halitosis beyond standardized counseling alone, but the zinc-containing mouthrinse produced the greatest short-term benefits across objective, clinician-rated, and patient-reported outcomes. These findings support zinc-based rinses as a practical short-term adjunct for managing persistent intra-oral halitosis in outpatient dental care. Durability after discontinuation and potential relapse beyond 4 weeks were not assessed in this trial. Full article

Background and Objectives: Halitosis can impair psychosocial well-being, and orthodontic appliances may modify plaque retention and oral ecology. We compared patient-perceived halitosis burden, clinician-rated malodor, and oral health-related quality of life (OHRQoL) among clear aligner users, fixed-brace patients, and untreated controls, and explored oral and salivary correlates of worse malodor severity. Methods: This cross-sectional study (March 2024–November 2025) enrolled 184 participants aged 15–35 years (aligners n = 62; fixed braces n = 64; controls n = 58). Outcomes were HALT (0–100), organoleptic score (0–5), and OHIP-14 (0–56). Plaque index, gingival inflammation, tongue coating, and unstimulated salivary flow were recorded; low flow was defined as <0.25 mL/min. Organoleptic score ≥ 2 was used descriptively for clinically relevant malodor prevalence, whereas organoleptic score ≥3 defined a moderate-to-severe malodor phenotype for secondary exploratory internal modeling. Multivariable robust linear models (HALT) and proportional-odds ordinal logistic regression (organoleptic severity) were used. Results: Fixed braces showed higher HALT (53.7 ± 6.2) than controls (46.3 ± 6.4) and aligners (41.7 ± 7.4) (p < 0.001), higher organoleptic scores (2.9 ± 0.4 vs. 2.4 ± 0.6 vs. 2.2 ± 0.6; p < 0.001), and worse OHIP-14 (18.6 ± 4.7 vs. 15.9 ± 4.3 vs. 13.8 ± 4.8; p < 0.001). Clinically relevant malodor prevalence (organoleptic ≥ 2) was 96.9% in fixed braces, 79.3% in controls, and 66.1% in aligners (p < 0.001); because ≥2 was used as a broad descriptive threshold, these values should be interpreted as descriptive rather than diagnostic prevalence estimates. In adjusted models, greater tongue coating, higher plaque, and low salivary flow were associated with worse organoleptic severity, whereas appliance category did not remain independently associated with HALT once concurrent clinical correlates were included. Conclusions: Fixed braces showed higher unadjusted malodor burden and worse OHRQoL than aligners and untreated controls, but appliance category should be interpreted as a contextual exposure linked to plaque-retentive conditions rather than as a standalone causal determinant. Plaque accumulation, tongue coating, and lower salivary flow showed the strongest associations with worse malodor severity. These findings should be interpreted in light of the cross-sectional design, possible observer and selection bias, and residual confounding. Full article

Diagnosis and monitoring of type 2 diabetes mellitus (T2DM) typically rely on invasive blood-based biomarkers. To explore non-invasive alternatives, this study examined tongue coating metabolites to identify metabolic signatures linked to diabetes progression. A case-control observational study categorized participants into control, prediabetes, and diabetes groups. Tongue coating samples were analyzed using liquid chromatography-mass spectrometry (LC-MS). Differential metabolites were correlated with clinical parameters, including HbA1c, BMI, and eGFR. Distinct metabolic profiles emerged across groups, with significant differences in five endogenous metabolites (phenylpyruvic acid, propionylcarnitine, pyridoxal 5′-phosphate, phenethylamine, phenethylamine glucuronide) and four amino acids (isoleucine, lysine, phenylalanine, tyrosine). Diabetic subjects showed elevated phenylpyruvic acid and phenethylamine, while propionylcarnitine, pyridoxal 5′-phosphate, and phenethylamine glucuronide were reduced. Phenethylamine was positively correlated with HbA1c; propionylcarnitine and phenethylamine glucuronide showed negative correlations with HbA1c and BMI. Detected total amino acids were inversely correlated with eGFR. Additionally, a diabetes index derived from these metabolic features also holds potential for discriminating disease states. These findings underscore the potential of tongue coating metabolites as a relatively non-invasive approach for evaluating T2DM states. The observed metabolic alterations provide valuable insights into diabetes-associated dysregulation, including protein glycation, obesity-related metabolic shifts, and renal impairment. Full article

Background/Objectives: Oral microbial homeostasis is crucial for overall health. Nonetheless, the relationship between probiotics and the oral environment remains unclear. This study investigated the association between continuous consumption of yogurt containing Lactobacillus delbrueckii ssp. bulgaricus OLL1073R-1 (LbR1 yogurt), salivary antimicrobial proteins (AMPs), tongue-coating microbiota, and upper respiratory tract infection (URTI) frequency. Methods: This observational study was conducted on 53 nursing home care workers, categorized into a group who consumed LbR1 yogurt daily for over 1 year (n = 40, yogurt group) and a non-intake group (n = 13, non-yogurt group). Salivary and tongue-coating samples were collected. Results: The yearly URTI frequency was lower in the yogurt group than in the non-yogurt group (p = 0.003). The salivary β-defensin-2 (HBD2) and β-defensin-3 (HBD3) flow rates were higher in the yogurt group than in the non-yogurt group (p = 0.02 and p = 0.0009, respectively). Fusobacterium nucleatum (F. nucleatum) ssp. animalis abundance was lower in the yogurt group than in the non-yogurt group (p = 0.04). Bayesian network analysis indicated an association between yogurt consumption and the yearly URTI frequency and salivary HBD2 and HBD3 flow rates. Conclusions: Continuous consumption of LbR1 yogurt was associated with elevated salivary HBD2 and HBD3, reduced abundance of F. nucleatum ssp. animalis, and decreased URTI frequency. Thus, LbR1 yogurt intake is associated with modulated oral immunity and microbiota, suggesting a potential link to reduced URTIs. However, as an observational pilot study, its results should be interpreted with caution. Full article

Background: Coronavirus disease 2019 (COVID-19) vaccines deliver mRNA packaged in lipid nanoparticles via intramuscular injection. This study investigated several factors influencing antibody production patterns and adverse reactions after vaccination with COVID-19 vaccines. Methods: Among the participants of the Iwaki Health Promotion Project (IHPP), 211 individuals who consented to this study were surveyed regarding antibody titers and adverse reaction symptoms following vaccination. A machine learning approaches such as ridge regression, elastic-net, light gradient boosting, and neural network were applied to extract the variables, and Bayesian network analysis was applied to explore causal relationships between health data and the multi-omics dataset obtained from the IHPP health checkups. Results: Females with lower levels of free testosterone experienced more adverse reactions than males. Moreover, the immune system is more active in younger individuals, causing adverse reactions and higher antibody production. The Spikevax vaccine induced adverse reaction symptoms with higher antibody production in cases of fever. Meanwhile, drinking 2–3 cups of green tea daily seemed to be effective in increasing antibody production. Factors increasing side effect risk include blood natural killer cell count and muscle quality in the vaccinated arm. Plasma metabolome metabolite concentrations, tongue coating bacterial colonization, and folate intake were also identified as factors influencing side effect risk. Furthermore, characteristics of participants at risk for fever symptoms included longer telomere length, higher antibody production patterns, and higher CD4-positive T cell counts. Conclusions: Further investigation of these identified influencing factors is expected to clarify the rationale for new vaccine development and identify lifestyle and dietary habits that enhance vaccine efficacy. Full article

Tongue coating is a crucial diagnostic indicator in traditional Chinese medicine, intuitively reflecting the body’s physiological and pathological conditions. However, traditional visual inspection methods are highly susceptible to subjective bias, often resulting in diagnostic deviations and inconsistencies. To address these limitations, this study proposes an intelligent tongue coating diagnostic model based on an enhanced YOLOv13. The model integrates a hybrid architecture of swin transformer and YOLOv13, effectively capturing global contextual and local textural features for fine-grained recognition and analysis of tongue coating characteristics. Experimental results show that the enhanced model substantially outperforms the original YOLOv13 in fine-grained feature extraction and boundary localization, establishing a reliable foundation for the objectification, standardization, and intelligent advancement of tongue diagnosis in traditional Chinese medicine. Full article

The oral cavity contains several microbial niches, including saliva, dental plaque and tongue coating, each shaped by distinct local environments and host factors. This study compared the ecological and functional characteristics of the microbiomes of these three oral sites within the same individuals and examined host conditions associated with their variation. Saliva, supragingival plaque and tongue coating samples were collected simultaneously from 31 adults without clinical oral lesions. The bacterial 16S rRNA gene (V3–V4 region) was sequenced using the Illumina MiSeq platform, and analyses included α and β diversity, Mantel correlations, differential abundance tests, network analysis and functional prediction. The three sites displayed a clear ecological gradient. Saliva and tongue coating were taxonomically similar but were influenced by different host factors, whereas plaque maintained a distinct, biofilm-like structure with limited systemic influence. Functional divergence was most pronounced on the tongue coating despite its taxonomic similarity to saliva, whereas functional differences between saliva and plaque were modest despite larger taxonomic separation. These findings indicate that microbial composition and function vary independently across oral niches and support the need for multi-site sampling to more accurately characterize oral microbial ecology. Full article

Background and Objectives: Morning halitosis undermines social well-being, yet the combined influence of basal salivary flow and tongue coating in healthy adults is unclear. Methods: In a cross-sectional study of 92 university students (18–35 years), we measured unstimulated salivary flow rate (uSFR), tongue-coating index (TCI), total volatile sulfur compounds (VSCs; Halimeter^®), organoleptic score (0–5), and self-perceived halitosis (yes/no) under standardized early-morning conditions. Results: Thirty-seven participants (40.2%) reported morning halitosis and showed lower uSFR (0.2 ± 0.1 vs. 0.3 ± 0.1 mL·min⁻¹) and higher TCI (2.3 ± 0.5 vs. 1.9 ± 0.4), with higher organoleptic scores (3.4 ± 0.6 vs. 2.1 ± 0.7) and VSCs (272.9 ± 39.8 vs. 163.7 ± 45.9 ppb; all p < 0.001). VSCs correlated inversely with uSFR (ρ = −0.58) and positively with TCI (ρ = 0.44). In multivariable models, uSFR (β = −0.53) and TCI (β = 0.31) explained 54% of VSC variance; each 0.1 mL·min⁻¹ fall in uSFR increased the odds of self-perceived halitosis 1.9-fold (p = 0.001). Conclusions: Even among healthy young adults, lower basal saliva and heavier tongue coating are independent contributors to morning malodor. Hydration, daily tongue cleaning, and addressing mouth breathing are pragmatic, first-line strategies. Full article

Background: Poststroke dysphagia is a common condition that can lead to complications such as aspiration pneumonia and malnutrition, significantly affecting the quality of life. Most patients recover their swallowing function spontaneously, but in others difficulties persist beyond six months. Can we predict this in advance? On the other hand, there have been recent attempts to use machine learning to predict disease prognosis. Therefore, this study aims to investigate whether machine learning can predict the long-term prognosis for poststroke dysphagia using early videofluoroscopic swallowing study (VFSS) data. Methods: Data from VFSSs performed within 1 month of onset and swallowing status at 6 months were collected retrospectively in patients with dysphagia who experienced their first acute stroke at a university hospital. We selected 14 factors (lip closure, bolus formation, mastication, apraxia, tongue-to-palate contact, premature bolus loss, oral transit time, triggering of pharyngeal swallow, vallecular residue, laryngeal elevation, pyriform sinus residue, coating of the pharyngeal wall, pharyngeal transit time, and aspiration) from the VFSS data, scored them, and analyzed whether they could predict the long-term prognosis using five machine learning algorithms: Random forest, CatBoost classifier, K-neighbor classifier, Light gradient boosting machine, Extreme gradient boosting. These algorithms were combined through an ensemble method to create the final model. Results: In total, we collected data from 448 patients, of which 70% were used for training and 30% for testing. The final model was evaluated using accuracy, precision, recall, F1-score, and Area Under the Receiver Operating Characteristic Curve (AUC), resulting in values of 0.98, 0.94, 0.84, 0.88, and 0.99, respectively. Conclusions: Machine learning models using early VFSS data have shown high accuracy and predictive power in predicting the long-term prognosis of patients with poststroke dysphagia, and they are likely to provide useful information for clinicians. Full article

Background: Halitosis is frequently observed in patients undergoing orthodontic treatment with multibracket appliances, primarily due to volatile sulfur compounds (VSCs) produced by oral anaerobic bacteria. Cetylpyridinium chloride (CPC) is a widely used antimicrobial agent in oral care products and may help alleviate halitosis.This study aimed to evaluate the effects of 0.05% CPC mouthwash on halitosis, oral hygiene indices, and the tongue microbiota in orthodontic patients with elevated VSC levels. Methods: In this randomized, double-blind, placebo-controlled clinical trial, 30 orthodontic patients with elevated VSCs (≥150 ppb) were assigned to a CPC mouthwash group or a placebo group. Participants used the assigned mouthwash three times daily for 1 month. Halitosis was quantitatively assessed by gas chromatography (Oral Chroma™), and oral hygiene parameters including Plaque Index (PI), Gingival Index (GI), Tongue Coating Index (TCI), and unstimulated salivary flow rate were evaluated at baseline and after the intervention. The tongue microbiota was analyzed by 16S rRNA sequencing. Results: The CPC mouthwash group showed significant reductions in total VSCs, hydrogen sulfide, methyl mercaptan, PI, GI, and TCI (p < 0.05), while salivary flow rate and dimethyl sulfide remained unchanged. Microbiome analysis revealed decreases in halitosis-associated genera (Actinomyces, Corynebacterium, Tannerella) and increases in beneficial species such as Streptococcus salivarius. Conclusions: CPC mouthwash (0.05%) effectively reduced halitosis and improved oral hygiene parameters in orthodontic patients, likely through modulation of the tongue microbiota. This mouthwash may serve as a safe and practical adjunct to conventional oral hygiene practices during orthodontic treatment. Full article

Background/Objectives: Coated tongue is a common oral condition with notable clinical relevance, often overlooked due to its asymptomatic nature. Its presence may reflect poor oral hygiene and can serve as an early indicator of underlying systemic diseases. This study aimed to develop a robust diagnostic model utilizing convolutional neural networks and machine learning classifiers to improve the detection of coated tongue lesions. Methods: A total of 200 tongue images (100 coated and 100 healthy) were analyzed. Images were acquired using a DSLR camera (Nikon D5500 with Sigma Macro 105 mm lens, Nikon, Tokyo, Japan) under standardized daylight conditions. Following preprocessing, feature vectors were extracted using CNN architectures (VGG16, VGG19, ResNet, MobileNet, and NasNet) and classified using Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Multi-Layer Perceptron (MLP) classifiers. Performance metrics included sensitivity, specificity, accuracy, and F1 score. Results: The SVM + VGG19 hybrid model achieved the best performance among all tested configurations, with a sensitivity of 82.6%, specificity of 88.23%, accuracy of 85%, and an F1 score of 86.36%. Conclusions: The SVM + VGG19 model demonstrated high accuracy and reliability in diagnosing coated tongue lesions, highlighting its potential as an effective clinical decision support tool. Future research with larger datasets may further enhance model robustness and applicability in diverse populations. Full article