IntroductionAccurate histopathological discrimination of lung adenocarcinoma (ACA) and squamous cell carcinoma (SCC) from benign tissue is a clinically demanding task in which the two malignant subtypes are frequently confused even by trained pathologists.MethodsWe present a dual-branch deep learning framework that couples a hierarchical shifted-window vision transformer (SwinV2-CR-Tiny) with a fully-convolutional masked-autoencoder-pretrained convolutional network (ConvNeXtV2-Tiny) to capture complementary global-contextual and local-textural representations of haematoxylin-and-eosin (H&E) stained tissue. The two branches are integrated through a learnable cross-feature gating module, refined by feature-space gated-modulation heads, and compressed by an Electric-Eel-Foraging-Optimizer (EEFO)-driven feature-selection stage. A Crested-Porcupine-Optimizer (CPO) calibrates class weighting and a Fennec-Fox-Optimizer (FFO) tunes per-class decision thresholds, while training combines Macenko stain normalization, CLAHE, MixUp/CutMix augmentation, a focal loss with label smoothing, Sharpness-Aware Minimization (SAM) and Stochastic Weight Averaging (SWA). Predictive uncertainty is quantified by Monte-Carlo (MC) dropout and model decisions are explained with Grad-CAM++ and Integrated Gradients. The model was evaluated on a stratified 70/15/15 partition of the LC25000 lung cohort (15,000 images; 5,000 per class).ResultsThe proposed model attains a test accuracy of 94.04%, macro-F1 of 0.9404, macro one-vs-rest AUC of 0.9886 and Cohen's κ of 0.9107, with near-perfect benign separation (sensitivity 0.999) and the residual error concentrated in the clinically expected ACA↔SCC confusion. Contrary to our prior expectation, MC-dropout predictive uncertainty did not separate misclassified from correctly classified cases on this cohort (mean across-pass standard deviation 0.0261 for errors versus 0.0269 for correct predictions). We additionally report a negative calibration finding: the model is systematically under-confident and naive temperature scaling increased the expected calibration error.DiscussionThe negative uncertainty result is attributed to the same confidence-suppressing training objective and is reported rather than omitted. We further discuss the data-provenance characteristics of LC25000 that constrain the interpretation of absolute accuracy. The framework offers an interpret-able, transparently-evaluated decision-support pipeline whose error profile aligns with established lung-cancer histomorphology.
BackgroundChronic inflammation is a recognized hallmark of cancer. The association between the inflammatory potential of diet and the risk of oral cancer remains unclear. We conducted a systematic review and meta-analysis to quantify this association.MethodsA comprehensive search of PubMed, EMBASE, and Web of Science was performed to identify observational studies investigating the relationship between dietary inflammatory potential scores and oral cancer risk. Pooled odds ratios (ORs) and 95% confidence intervals (CIs) were calculated using a random-effects model.ResultsSeven observational studies met the inclusion criteria. The meta-analysis revealed that individuals in the highest category of dietary inflammatory potential had a significantly higher risk of oral cancer compared to those in the lowest category (Pooled OR = 1.92, 95% CI: 1.41–2.62). The continuous analysis indicated that each 1-unit increment in the dietary inflammatory potential score was associated with a 19% increase in oral cancer risk. Sensitivity analysis using the leave-one-out approach confirmed the robustness of the pooled estimates.ConclusionOur pooled analysis supports an association between dietary inflammatory potential and oral cancer risk. These findings suggest that dietary interventions aimed at reducing inflammation may be a viable strategy for oral cancer prevention.
BackgroundPostoperative recurrence remains a major challenge in early-stage non–small cell lung cancer (NSCLC), and pathological TNM staging does not fully capture within-stage heterogeneity. We aimed to develop and externally validate an explainable machine learning model for recurrence prediction after curative resection.MethodsThis retrospective study included 723 patients with stage I–II NSCLC, including 52 recurrences, with independent external validation in a separate cohort (n=50). A Random Forest model using routinely available clinical and pathological variables was developed within a nested cross-validation framework and compared with logistic regression. Performance was evaluated using ROC-AUC, calibration, Decision Curve Analysis, and SHAP-based interpretation.ResultsThe Random Forest achieved a mean internal ROC-AUC of 0.70 versus 0.64 for logistic regression, although no formal paired statistical comparison was performed. In an independent case-control external validation cohort with an artificially balanced outcome distribution, the Random Forest achieved an ROC-AUC of 0.68 (95% CI 0.53–0.83); the balanced design precluded assessment of calibration or absolute risk at natural prevalence. Sigmoid recalibration of pooled out-of-fold predictions yielded an apparent Brier score reduction to 0.063, although post-calibration performance was not independently evaluated. Decision Curve Analysis demonstrated positive net clinical benefit across clinically relevant thresholds. SHAP analysis highlighted tumor size, pathological T stage, and STAS among the principal tumor-related predictors, whereas fold-level analysis showed greater stability for tumor size, tumor necrosis, and STAS. Complementary time-to-event analyses accounting for right censoring included 715 patients and 44 recurrence events.ConclusionAn explainable machine learning model based on routinely available variables demonstrated promising but preliminary discrimination in an independent case-control external validation cohort. Larger consecutive cohorts with natural outcome prevalence are required before clinical application.
ObjectiveThe systematic review and meta-analysis aimed to evaluate the effects of combined aerobic and resistance exercise (CE) on metabolic blood biomarkers (MBB), quality of life (QoL), and QoL subscales in women with breast cancer (BC). A secondary objective was to investigate whether exercise dose characteristics influenced these outcomes.MethodsRandomized controlled trials (RCTs) were comprehensively searched for studies investigating the effects of CE on BC patients across four databases (PubMed, Cochrane, MEDLINE, and Web of Science). The standardized mean difference (Hedges’ g) was calculated using Comprehensive Meta-Analysis (CMA V4), and the risk of bias was assessed using RevMan 5.4.ResultsOf the 2,654 records identified, 37 RCTs met the inclusion criteria, with 24 RCTs focusing on overall QoL, nine evaluating FACT-B, and six to nine RCTs assessing MBB outcomes. CE exercise significantly improved overall QoL (Hedges’ g = 1.10; 95% CI: 0.72, 1.48; p < 0.0001), FACT-B (Hedges’ g = 1.14; 95% CI: 0.46, 1.81; p = 0.001), HDL (Hedges’ g = 0.828; 95% CI: 0.051, 1.606; p = 0.037), HOMA-IR (Hedges’ g = − 0.581, 95% CI: − 1.059, − 0.103; p = 0.017), and DBP (Hedges’ g = − 0.39; 95% CI: − 0.786, − 0.0008; p = 0.045). No significant effects were observed for glucose, insulin, triglycerides, LDL, or SBP. Secondary analyses further suggested that higher exercise doses may be associated with greater improvements in these outcomes than shorter exercise doses.ConclusionCE was associated with significant improvements in MBB, QoL, and QoL subscales in breast cancer patients; however, substantial heterogeneity should be considered when interpreting these findings. Exercise dose characteristics may influence QoL and selected MBB outcomes. Further studies are needed to evaluate the optimal dose–response effects of CE in BC patients.Systematic Review Registrationhttps://www.crd.york.ac.uk/PROSPERO/view/CRD420251032722, identifier CRD420251032722.
