Mehmet Ali Cengiz

High-temperatures significantly constrain the growth and resilience of day-neutral strawberry ( Fragaria × ananassa ) varieties, affecting their morphological, physiological, and biochemical characteristics. In this study, frigo plants of six day-neutral strawberry cultivars (Albion, Monterey, Portola, Redlands Hope, San Andreas, and Sweet Ann) were initially grown for 5 weeks in 15 cm diameter plastic pots (3:1:1; garden soil, peat, perlite) under open conditions. After 2–3 weeks of acclimatization at 23/15 °C (day/night) in a climate chamber, the temperature in one chamber was maintained at 23/15 °C, while the other was subjected to high-temperature conditions at 30/25 °C. Temperature and cultivars significantly influenced most of the morphological, physiological, and biochemical characteristics of strawberries. High- temperature caused a significant decrease in root length, number of leaves, leaf area, leaf fresh and dry weight, root fresh and dry weight, plant fresh and dry weight, and membrane stability index, while increasing leaf surface temperature, Malondialdehyde content, starch content, sugar content, and total carbohydrate content of strawberry plants, indicating enhanced oxidative stress and metabolic reprogramming. Among varieties, the highest values of the number of leaves, leaf area, leaf fresh and dry weight, root fresh weight, and plant fresh and dry weight were recorded in the San Andreas variety. The Portola variety exhibited considerably higher values of root length and root dry weight. The leaf surface temperature was higher in Monterey, and the membrane stability index was higher in San Andreas. The highest values of MDA and sugar content were observed in the Albion strawberry variety, and the highest values of starch and total carbohydrate content were observed in Monterey. The findings highlight the detrimental effects of high-temperatures on day-neutral strawberry varieties and emphasize the importance of selecting heat-tolerant cultivars for sustainable strawberry production under changing climatic conditions.

Environmental stress factors, especially salinity, are among the most important abiotic stresses that negatively affect plant production worldwide. High salt levels in irrigation water are a major abiotic stress factor that significantly reduces spinach physiological functioning and production, particularly in irrigated areas. Improving the salt tolerance of spinach is critical for sustainable production, and in this study, we tested the hypothesis that exogenous proline (5 µM), ascorbic acid (1 mM), and salicylic acid (1 mM) applications, applied separately, would reduce salinity stress. These applications were performed at regular 14-day intervals starting from the third true leaf stage. For this purpose, plants were exposed to irrigation water salinities of 0.38, 2.0, 4.0, 7.0, 10.0, and 15.0 dS m⁻¹, and growth, photosynthetic performance, antioxidant enzyme activities, lipid peroxidation, endogenous proline, and mineral contents were assessed. Increasing salinity to 15 dS m⁻¹ decreased leaf area by 53.23% and stomatal conductance by 83.07%, and all these physiological changes were statistically significant. Under salinity conditions, catalase, guaiacol peroxidase, glutathione reductase, glutathione S-transferase, and superoxide dismutase activities increased by 1.13–2.52-fold, while ascorbate peroxidase activity decreased by 59.69%. Malondialdehyde levels increased 6-fold with salinity, indicating enhanced oxidative damage. Consequently, yield decreased by 31% under 15 dS m⁻¹ salinity. Although all exogenous applications alleviated salinity stress, the most significant improvement was observed in proline application. Proline increased yield and chlorophyll content by 9% and 8.5%, respectively, and also increased antioxidant enzyme activities by 24.4–66.7%. Salicylic acid treatment increased the K⁺/Na⁺ ratio by 26.6%, and ascorbic acid treatment increased the Ca²⁺/Na⁺ ratio by 36.6%. Overall, low-dose proline application was found to improve photosynthetic pigment content and stomatal conductance, antioxidant defenses, and ion homeostasis in spinach against salinity stress, providing a stronger protective effect compared to ascorbic acid and salicylic acid. Furthermore, it can be concluded that proline application could be an effective way to manage salinity-induced limitations to physiological processes and yields, providing practical applications for sustainable production under saline irrigation conditions.

Feature weighting plays a central role in medical classification by enhancing predictive accuracy, interpretability, and clinical trust through the explicit quantification of variable relevance. Despite their widespread use, existing filter-, wrapper-, and embedded-based feature weighting methods are predominantly deterministic and exhibit pronounced sensitivity to label noise and outliers, which are pervasive in real-world medical data. This often results in unstable importance estimates and unreliable clinical interpretations. In this work, we introduce a novel Bayesian feature weighting model that fundamentally departs from existing approaches by jointly integrating simplex-constrained Dirichlet priors for global feature weights, hierarchical shrinkage priors for coefficient regularization, and contamination-aware priors for explicit modeling of label noise within a single coherent probabilistic framework. Unlike conventional Bayesian feature selection or robust classification models, the proposed formulation yields globally interpretable feature weights defined on the probability simplex, while simultaneously providing full posterior uncertainty quantification and robustness to both mislabeled observations and aberrant feature values through principled influence control. Comprehensive simulation studies across diverse contamination scenarios, together with applications to multiple real-world medical datasets, demonstrate that the proposed model consistently outperforms classical and state-of-the-art baselines in terms of discrimination, probabilistic calibration, and stability of feature-importance estimates. These results highlight the practical and methodological significance of the proposed framework as a robust, uncertainty-aware, and interpretable solution for medical decision making under noisy data conditions.

Purpose: The study aims to provide reference ranges of complete blood count (CBC) parameters in healthy children in Türkiye. Methods: Children aged 1 month to −18 years who were administered to the Well Child Outpatients Clinics of Department/Unit of Social Pediatrics in 10 centers in different regions of Türkiye were included in the study. Complete blood count measurements were collected from these centers. Results: A total of 24 115 healthy infants and children (12 204 males and 11 911 females) aged 1 month–18 years old were enrolled. Mean and median values of the red blood cell, white blood cell, and platelet parameters according to the age groups are presented in this study. In addition, the nonparametrically determined CBC reference intervals (RIs) corresponding to the 2.5th to 97.5th percentile, stratified by age and sex were provided. Conclusion: This is the first study that established RIs for hematological parameters in Turkish children younger than 18 years old. These data provide a more reliable basis for hematological evaluation across all pediatric healthcare disciplines in Türkiye.

Accurate outcome prediction often requires modeling complex interactions between input features and context-specific modifiers. The pliable lasso is a flexible regression framework that integrates such modifiers into the prediction process. In many real-world applications, however, these modifiers are unobserved at test time and must be estimated. This study investigates the performance of eight supervised machine learning algorithms for estimating the modifier matrix Z in a pliable lasso model under a known-to-unknown scenario. The analysis considers both classification accuracy for modifier estimation and regression accuracy for the final response prediction, using simulated data and two relevant real-world datasets: the Superconductivity dataset and the Mice Protein Expression dataset. Results indicate that tree-based ensemble models (e.g., XGBoost, Random Forest, and Decision Tree) deliver superior modifier classification (AUC > 0.99), while regularized models such as Lasso and Elastic Net achieve the best regression performance. The findings support a hybrid modeling approach in which tree-based classifiers estimate modifying variables, followed by regularized regression for accurate and interpretable predictions.

Background/Objectives: Fetal health is essential in prenatal care, influencing both maternal and fetal outcomes. Cardiotocography (CTG) monitors uterine contractions and fetal heart rate, yet manual interpretation exhibits significant inter-examiner variability. Machine learning offers automated alternatives; however, class imbalance in CTG datasets where pathological cases constitute less than 10% leads to poor detection of minority classes. This study aims to provide the first systematic benchmark comparing five resampling strategies across seven classifier families for multi-class CTG classification, evaluated using imbalance-aware metrics rather than overall accuracy alone. Methods: Seven machine learning models were employed: Naïve Bayes (NB), Random Forest (RF), Linear Discriminant Analysis (LDA), k-Nearest Neighbors (KNN), Linear Support Vector Machine (SVM), Multinomial Logistic Regression (MLR), and Multi-Layer Perceptron (MLP). To address class imbalance, we evaluated the original unbalanced dataset (base) and five resampling methods: SMOTE, BSMOTE, ADASYN, NearMiss, and SCUT. Performance was evaluated on a held-out test set using Balanced Accuracy (BACC), Macro-F1, the Macro-Matthews Correlation Coefficient (Macro-MCC), and Macro-Averaged ROC-AUC. We also report per-class ROC curves. Results: Among all models, RF proved most reliable. Training on the original distribution (base) yielded the highest BACC (0.9118), whereas RF combined with BSMOTE provided the strongest class-balanced performance (Macro-MCC = 0.8533, Macro-F1 = 0.9073) with a near-perfect ROC-AUC (approximately 0.986–0.989). Overall, resampling effects proved model dependent. While some classifiers achieved optimal performance on the natural class distribution, oversampling techniques, particularly SMOTE and BSMOTE, demonstrated significant improvements in minority class discrimination and class-balanced metrics across multiple model families. Notably, certain models benefited substantially from resampling, exhibiting enhanced Macro-F1, BACC, and minority class recall without sacrificing overall accuracy. Conclusions: These findings establish robust, model-agnostic baselines for CTG-based fetal health screening. They highlight that strategic oversampling can translate improved minority class discrimination into clinically meaningful performance gains, supporting deployment in cost-sensitive and threshold-aware clinical settings.

Real-world systems frequently exhibit hierarchical multipartite graph structures, yet existing graph neural network (GNN) approaches lack systematic frameworks for hyperparameter optimization in heterogeneous multi-level architectures, limiting their practical applicability. This study proposes a Bayesian optimization framework specifically designed for heterogeneous GNNs operating on three-level graph structures, addressing the computational challenges of configuring partition-aware architecture. Four GNN architectures—Graph Convolutional Networks (GCNs), Graph Attention Networks (GATs), Graph Isomorphism Networks (GINs), and GraphSAGE—were systematically evaluated using Gaussian Process-based Bayesian hyperparameter optimization with inter-partition message-passing mechanisms. The framework was validated on the TIMSS 2023 dataset (10,000 students, 789 schools, 25 countries), demonstrating that Bayesian-optimized GraphSAGE achieved the highest explained variance (R² = 0.6187, RMSE = 71.73, MAE = 64.32) compared to seven baseline methods. Bayesian optimization substantially improved model performance, revealing that two-layer architectures optimally capture cross-partition dependencies in three-level structures. GNNExplainer was used to identify the most influential student-level features learned by the model, providing explanatory insight into how the model represents individual characteristics. The optimization framework is broadly applicable to other heterogeneous and multi-level graph settings; however, the empirical findings, such as the optimal architecture depth, are specific to hierarchical graphs with structural properties like the TIMSS topology.