Metformin remains the cornerstone of pharmacological management for Type 2 diabetes mellitus (T2DM); however, its long-term use has been associated with impaired vitamin B12 absorption and macrocytosis. The prevalence and independent determinants of macrocytosis in metformin-treated populations remain insufficiently characterized. A cross-sectional observational study was conducted at Saidu Teaching Hospital, Swat, between July 2024 and August 2025. A total of 236 adults with T2DM receiving metformin therapy for at least six months were enrolled. Macrocytosis was defined as a mean corpuscular volume (MCV) exceeding 100 fL. Demographic characteristics, clinical parameters, and concomitant medication use were systematically recorded. Univariate analyses were initially performed, followed by multivariate logistic regression to identify independent predictors of macrocytosis. The overall prevalence of macrocytosis was 16.1% (95% confidence interval (CI): 11.7–21.4%). Multivariate analysis demonstrated that metformin therapy duration >4 years (odds ratio (OR): 3.42, p = 0.002), daily metformin dose ≥1500 mg (OR: 2.89, p = 0.006), concomitant proton pump inhibitor (PPI) use (OR: 4.15, p < 0.001), and age ≥60 years (OR: 2.26, p = 0.030) were independently associated with macrocytosis, with concurrent PPI use emerging as the strongest predictor. These findings indicate that macrocytosis is a relatively common hematologic abnormality in metformin-treated adults with T2DM and is strongly influenced by treatment duration, dosage intensity, age, and concurrent PPI therapy. Risk-stratified surveillance strategies incorporating periodic assessment of MCV and serum vitamin B12 levels may therefore be warranted, particularly in high-risk patients, to enhance patient safety and optimize the long-term clinical management of T2DM.

Malignant mesothelioma remains a diagnostic challenge due to the phenotypic overlap with benign pleural diseases and the reliance on invasive procedures for definitive confirmation. To address these limitations, a leakage-aware, explainable machine learning framework was developed and applied to a publicly available mesothelioma dataset comprising 324 cases (96 mesothelioma, 228 symptomatic non-mesothelioma). Variables prone to target leakage or unavailable at the point of diagnosis—such as diagnosis method, cytology results, mesothelioma subtype, and survival status—were systematically excluded. The remaining features were stratified into initial clinical presentation, post-imaging, and post-pleural-fluid analysis stages prior to model development. The dataset was partitioned into a development cohort (n = 226) and an independent hold-out cohort (n = 98). Multiple classifiers, including logistic regression, support vector machine, k-nearest neighbors, and light gradient boosting machine, were optimized via grid search and evaluated using repeated stratified 5-fold cross-validation. The diagnosis method was identified as a perfect inverse surrogate of the target variable and consequently removed. The light gradient boosting machine exhibited superior performance, achieving the highest average precision (0.543) and Matthews correlation coefficient (0.306) during cross-validation. On the unseen hold-out cohort, light gradient boosting machine yielded an area under the receiver operating characteristic curve of 0.660, average precision of 0.483, balanced accuracy of 0.615, and Matthews correlation coefficient of 0.233. At the conventional 0.50 threshold, sensitivity was 0.448, specificity 0.783, and negative predictive value 0.771; lowering the threshold to 0.30 increased sensitivity to 0.690 at the expense of specificity reduction to 0.493. SHapley Additive exPlanations (SHAP) identified age, platelet count, lung side, white blood cell count, and duration of asbestos exposure as the most influential predictors. This leakage-aware, explainable light gradient boosting machine model delivers clinically interpretable diagnostic predictions while mitigating target leakage, demonstrating moderate discrimination and potential utility in real-world clinical settings. These findings warrant further external validation and prospective evaluation to confirm generalizability and clinical impact.

Traditional cross-sectional methods could not adequately capture time-to-event dynamics among HIV patients on antiretroviral therapy (ART). Survival analysis provides a more appropriate framework for examining predictors of clinical outcomes. This study applied a multivariable Cox proportional hazards model to examine socio-demographic and clinical factors associated with time to treatment interruption or death among 2,887 HIV patients on ART at Specialist Hospital Yola, Adamawa State, Nigeria. The outcome was time to treatment interruption or death, with Current ART Status serving as the failure indicator (1 = interruption in treatment (IIT) or dead; 0 = censored). Independent variables were age group, sex, marital status, and viral load status (suppressed <1000 copies/ml versus unsuppressed ≥1000 copies/ml). Kaplan-Meier curves, Log-rank tests, and Cox regression models were employed in the investigation. The proportional hazards assumption was assessed using log-minus-log plots, Schoenfeld residuals, and the global Schoenfeld test. Most participants were females (70.1%) and had suppressed viral load (95%). Unsuppressed viral load was significantly associated with treatment interruption or death in both univariate (hazard ratio (HR) = 1.98; 95% confidence interval (CI): 1.44–2.73; p < 0.001) and multivariate analyses (adjusted HR = 1.96; 95% CI: 1.42–2.70; p < 0.001). Age group, sex, and marital status were not significant predictors. The proportional hazards assumption was satisfied (global Schoenfeld test p = 0.669). Unsuppressed viral load was therefore the only independent predictor of treatment interruption or death. Strengthening viral load monitoring and adherence support should be prioritized to improve treatment outcomes.

Maternal health forms a pivotal public health concern since poor pregnancy outcomes often arise from delayed detection of physiological abnormalities. Recent advances in Internet of Things (IoT)-based maternal health monitoring devices have enabled automated risk assessment of real-time clinical data. This study developed a Machine Learning (ML) framework to categorize maternal health risk into High, Medium, and Low Levels using a dataset of 1,014 records collected from maternal health centres. Six physiological indices, including body temperature, heart rate, systolic, and diastolic blood pressure, blood glucose level, and age were analyzed via five-fold cross validation. Logistic Regression (LR), Decision Tree (DT), and Random Forest (RF) models were trained and evaluated. RF model achieved the best performance of 0.88 accuracy and F1-score of 0.88. Taken together, recall of 0.91 for high-risk cases, area under the curve (AUC) of 0.973, log loss of 0.449, mean squared error (MSE) of 0.228, root mean squared error (RMSE) of 0.478, and mean absolute error (MAE) of 0.157 demonstrated reliable prenatal risk prediction despite limited resources in healthcare settings.