Automated essay scoring (AES) systems have gained significant importance in educational settings, offering a scalable, efficient, and objective method for evaluating student essays. However, developing AES systems for Arabic poses distinct challenges due to the language’s complex morphology, diglossia, and the scarcity of annotated datasets. This paper presents a hybrid approach to Arabic AES by combining text-based, vector-based, and embedding-based similarity measures to improve essay scoring accuracy while minimizing the training data required. Using a large Arabic essay dataset categorized into thematic groups, the study conducted four experiments to evaluate the impact of feature selection, data size, and model performance. Experiment 1 established a baseline using a non-machine learning approach, selecting top-N correlated features to predict essay scores. The subsequent experiments employed 5-fold cross-validation. Experiment 2 showed that combining embedding-based, text-based, and vector-based features in a Random Forest (RF) model achieved an R2 of 88.92% and an accuracy of 83.3% within a 0.5-point tolerance. Experiment 3 further refined the feature selection process, demonstrating that 19 correlated features yielded optimal results, improving R2 to 88.95%. In Experiment 4, an optimal data efficiency training approach was introduced, where training data portions increased from 5% to 50%. The study found that using just 10% of the data achieved near-peak performance, with an R2 of 85.49%, emphasizing an effective trade-off between performance and computational costs. These findings highlight the potential of the hybrid approach for developing scalable Arabic AES systems, especially in low-resource environments, addressing linguistic challenges while ensuring efficient data usage.

Automated Essay Scoring (AES) in Arabic remains underdeveloped due to the language's diglossia, rich morphology, and limited annotated resources, reflecting inherent linguistic asymmetry. This study advances Arabic AES by introducing a BERT-based sentence-pair cross-encoder framework that leverages the structural symmetry of bidirectional attention and combines it with an Integrated Gradients (IG) explanation pipeline. Five widely used Arabic BERT variants were systematically evaluated: (1) asafaya/bert-base-arabic, (2) aubmindlab/arabertv02, (3) UBC-NLP/MARBERT,faisalq/ (4) SaudiBERT, and (5) CAMeL-Lab/CAMeLBERT-MSA. Across the experiments, the CAMeLBERT-MSA cross-encoder fine-tuned for ten epochs achieved high results, with an R2 of 98.47%, MAE of 0.07, and 98.32% accuracy within ± 0.5 points. This marks a significant improvement over both shallow similarity-based baselines and earlier BERT concatenation models, demonstrating the value of deeper task-specific adaptation on Modern Standard Arabic (MSA) corpora. Cross-prompt evaluations revealed both generalization symmetry-strong performance with limited prompts (R2 = 79.38%, 86.92% accuracy within ± 1.0 point at four prompts)-and generalization asymmetry, where performance declined as prompt heterogeneity increased. The interpretability pipeline enhanced pedagogical value by producing student-only, word-level rationales that identified rubric-aligned, score-raising, and score-lowering terms. These lexicons serve as lightweight concept inventories, reinforcing fairness and trust through explanatory symmetry between model reasoning and rubric expectations.

Background: Measuring glycated hemoglobin (HbA1c) is essential for assessing long-term glycemic control, yet direct testing remains expensive and underutilized in many large-scale health surveys and resource-constrained settings. This study aims to (i) deliver a highly accurate and interpretable ML model for predicting HbA1c from routinely collected clinical, biochemical, and demographic data, (ii) reduce dependency on extensive laboratory panels by identifying a compact, cost-efficient subset of key predictors, and (iii) establish a transferable, explainable modeling framework applicable across chronic disease biomarkers. Unlike prior HbA1c prediction studies that focus primarily on classification or accuracy-driven models, this work introduces a unified framework for continuous HbA1c regression that jointly integrates cost-oriented feature parsimony, stratified regression validation, and explainability by design. Methods: We aggregated data from the National Health and Nutrition Examination Survey (NHANES) cycles 2007-2020, encompassing 66,148 records and 224 candidate features. We implemented a two-stage feature selection pipeline: Incremental Correlation Selection (ICS) to narrow the variable space, followed by Recursive Feature Elimination with Cross-Validation (RFECV) to isolate the most informative features. Model interpretability was assessed using partial dependence plots and feature importance analysis. Results: The optimal model, LightGBMRegressor with most-frequent imputation, achieved R2 = 0.7161, MAE = 0.334, MSE = 0.304, and MAPE = 5.56%, while using only 40 selected features. Interpretability analysis revealed clinically coherent relationships that align with physiological expectations. Discussion: The proposed framework maintains robust predictive performance while substantially reducing the number of required input features, enabling cost-efficient HbA1c estimation together with transparent, physiologically coherent model insights. By consolidating continuous HbA1c prediction, cost-aware feature selection, stratified evaluation, and explainability within a single pipeline are enhanced. Conclusions: This study advances beyond existing approaches and offers a practical blueprint for scalable biomarker estimation in population health and clinical decision-support applications. Its explainable, efficient, and generalizable design positions it as a strong candidate for clinical decision-support and population-health applications.

The sustainable management of water resources and the development of climate-resilient infrastructure depend on the precise identification of water bodies in satellite imagery. This paper presents a novel deep learning architecture that integrates a convolutional block attention module (CBAM) into a modified EfficientNet–UNet backbone. This integration allows the model to prioritize informative features and spatial areas. The model robustness is ensured through a rigorous training regimen featuring five-fold cross-validation, dynamic test-time augmentation, and optimization with the Lovász loss function. The final model achieved the following values on the independent test set: precision = 90.67%, sensitivity = 86.96%, specificity = 96.18%, accuracy = 93.42%, Dice score = 88.78%, and IoU = 79.82%. These results demonstrate improvement over conventional segmentation pipelines, highlighting the effectiveness of attention mechanisms in extracting complex water-body patterns and boundaries. The key contributions of this paper include the following: (i) adaptation of CBAM within a UNet-style architecture tailored for remote sensing water-body extraction; (ii) a rigorous ablation study detailing the incremental impact of decoder complexity, attention integration, and loss function choice; and (iii) validation of a high-fidelity, computationally efficient model ready for deployment in large-scale water-resource and ecosystem-monitoring systems. Our findings show that attention-guided segmentation networks provide a robust pathway toward high-fidelity and sustainable water-body mapping.

This study focuses on optimizing C2 yields in the oxidative coupling of methane (OCM), a pivotal process for sustainable chemical production. By harnessing advanced machine learning (ML) techniques, this research aimed to predict C2 yields and identify the factors that drive catalytic performance. The Extra Trees Regressor emerged as the most effective model after a comprehensive evaluation across multiple datasets and methodologies. Key to the method was the use of an innovative Aggregated Catalyst Physicochemical Descriptor (ACPD) and stratified cross-validation, which effectively addressed feature complexity and target skewness. Hyperparameter optimization using Modified Sequential Model-Based Optimization (SMBO) further enhanced the model’s performance, achieving optimized R2 values of 61.7%, 75.9%, and 92.0% for datasets A, B, and C, respectively, with corresponding reductions in the Mean Squared Error (MSE) and Root Mean Squared Error (RMSE). Additionally, SHAP (SHapley Additive exPlanations) analysis provided a detailed understanding of the model’s decision-making process, revealing the relative importance of individual features and their contributions to the predictive outcomes. This research not only achieved state-of-the-art predictive accuracy, but also deepened our understanding of the underlying chemical dynamics, offering practical guidance for catalyst design and operational optimization. These findings mark a significant advancement in catalysis, paving the way for future innovations in sustainable chemical manufacturing.

Finger-vein recognition offers a highly secure and non-intrusive biometric modality ideal for modern authentication systems, yet its deployment on resource-limited devices remains challenging due to the high computational cost of deep models and the rigidity of single-finger enrollment. We present a new two-stage deep learning framework that solves these problems by combining a lightweight MobileNet feature extractor that gets better over time with a new Dense-Head Probabilistic Siamese (DHPS) matcher. Our method is the only one that allows for multi-finger recognition, so users can authenticate with any finger without losing accuracy or speed. Layer-wise unfreezing carefully adjusts the feature extractor to find the right balance between model compactness and discriminative power. The DHPS matcher replaces traditional margin-based losses with a calibrated probabilistic output that is optimized through binary cross-entropy. Our system gets the best results on three different public finger-vein datasets, with Equal Error Rates (EER) of 0.002, 0.067, and 0.075 on the FV-USM, UTFVP, and VERA datasets, respectively. It also has F1-scores of 99.8%, 95.6%, and 91.3% on their test sets. These results are possible because the model is small and can be used on embedded platforms with fast inference. In addition, the pre-trained extractor will be made available to the public to encourage more research in the future. This work improves practical finger-vein biometrics by making them more accurate, flexible, and efficient, which gets around major obstacles to their use in the real world.

In today’s rapidly advancing healthcare landscape, integrating Artificial Intelligence (AI) and Machine Learning (ML) has the potential to significantly improve patient care and streamline medical processes. The utilization of confidential patient data to train and develop these technologies, however, raises significant concerns regarding authenticity, security, and privacy. In this study, we introduce MediChainAI, a safe and practical framework that allows patients full ownership over their own health data by integrating Self-Sovereign Identity (SSI), Blockchain, and sophisticated cryptography techniques. By clearly outlining the goals and parameters of this access, MediChainAI allows patients to safely and selectively share data with healthcare providers and researchers. While SSI guarantees that patients have ownership of their data, the framework uses Blockchain technology to keep things transparent and secure. Further, MediChainAI makes use of Merkle trees, which provide verified access to subsets of data without jeopardizing the privacy of the whole dataset. The encryption mechanism, which is based on smart contracts, is a distinctive feature of the framework that allows researchers and medical practitioners controlled and secure access to patient data. In order to improve the accuracy and reliability of medical diagnoses and treatment, this strategy makes sure that only confirmed, legitimate data is utilized to train medical models. A significant step toward safer and more personalized healthcare, MediChainAI encourages ethical and patient-focused innovation by effectively resolving essential issues regarding data security and patient privacy.

Cardiovascular diseases are a global health challenge that necessitates improvements in diagnostic accuracy and efficiency. This study examines the potential of deep learning (DL) models for the classification of electrocardiogram (ECG) images to assist in the identification of various cardiac conditions. We initiated a two-tiered experimental framework to investigate the effectiveness of several neural network architectures in this medical application. In the first experiment, eight distinct neural network models were selected based on their top-5 accuracy on the ImageNet validation dataset and were fine-tuned using transfer learning techniques. These models were assessed using a cross-validation scheme, focusing on balanced accuracy, precision, recall, and the F1-score to evaluate their classification capabilities across four cardiac conditions: Myocardial Infarction (MI), abnormal heartbeat, historical MI, and normal ECG patterns. The second experiment extended our inquiry into the power of ensemble learning. By testing all possible combinations of the chosen models, we explored 120 ensemble configurations. The resulting analysis identified the best-performing ensemble set, which did not include the least effective model based on F1 score rankings. The most effective ensemble, composed of Inception, MobileNet, and NASNetLarge, achieved an F1 score of 0.9651 and a balanced accuracy of 0.9640, indicating a superior predictive performance. The ROC curve analysis yielded near-perfect Area Under the Curve (AUC) values for all classes, underscoring the ensemble’s proficiency in distinguishing between the specified cardiac conditions. The outcomes of this research highlight the synergistic benefit of ensembles in DL applications for medical imaging and suggest a promising approach for the early detection and diagnosis of cardiac diseases, potentially improving clinical outcomes and patient care.

Breast cancer significantly affects people's health globally, making early and accurate diagnosis vital. While ultrasound imaging is safe and non-invasive, its manual interpretation is subjective. This study explores machine learning (ML) techniques to improve breast ultrasound image segmentation, comparing models trained on combined versus separate classes of benign and malignant tumors. The YOLOv8 object detection algorithm is applied to the image segmentation task, aiming to capitalize on its robust feature detection capabilities. We utilized a dataset of 780 ultrasound images categorized into benign and malignant classes to train several deep learning (DL) models: UNet, UNet with DenseNet-121, VGG16, VGG19, and an adapted YOLOv8. These models were evaluated in two experimental setups-training on a combined dataset and training on separate datasets for benign and malignant classes. Performance metrics such as Dice Coefficient, Intersection over Union (IoU), and mean Average Precision (mAP) were used to assess model effectiveness. The study demonstrated substantial improvements in model performance when trained on separate classes, with the UNet model's F1-score increasing from 77.80 to 84.09% and Dice Coefficient from 75.58 to 81.17%, and the adapted YOLOv8 model achieving an F1-score improvement from 93.44 to 95.29% and Dice Coefficient from 82.10 to 84.40%. These results highlight the advantage of specialized model training and the potential of using advanced object detection algorithms for segmentation tasks. This research underscores the significant potential of using specialized training strategies and innovative model adaptations in medical imaging segmentation, ultimately contributing to better patient outcomes.

Multiple sclerosis (MS) is a chronic autoimmune disease characterized by inflammatory brain lesions, making MRI-based lesion segmentation challenging due to noise, missing data, and limited availability of high-quality labeled images. This paper presents RAUM-GANs, a multi-layer deep learning framework designed to address these challenges and enhance segmentation accuracy. The preprocessing stage comprises three layers: (1) noise reduction using a modified Denoising GAN (DGAN-Net), achieving peak signal-to-noise ratio (PSNR) values up to 42.21 dB across varying noise levels; (2) missing data imputation through advanced GAN-based methods, ensuring clinically reliable reconstruction of incomplete MRI scans; and (3) dataset expansion via a Multi-level Identity GAN (MGAN), which incorporates an identity block to prevent mode collapse, an 8-connected pixel constraint to maintain spatial coherence, and a softened discriminator output to mitigate vanishing gradients. For segmentation, a Residual Attention U-Net (RAU-Net) with identity mapping is employed, yielding precise detection and delineation of MS lesions. Extensive evaluation on the MICCAI MSSEG-2 dataset demonstrates that RAUM-GANs outperform four state-of-the-art methods, achieving a Dice score of 96.6%, Fréchet Inception Distance (FID) of 43.13, and Inception Score (IS) of 14.03. The results highlight the framework’s ability to generate high-quality synthetic MRI data, improve robustness against noise and incomplete information, and deliver superior lesion segmentation performance. RAUM-GANs provides a comprehensive, scalable solution for MS lesion analysis, with potential applicability to other medical imaging domains where data quality and scarcity remain significant barriers.

What if you could identify someone's right palmprint just by looking at their left-and vice versa? That is exactly what I set out to do. I built a specially adapted Siamese network that only needs one palm to reliably recognize the other, making biometric systems far more flexible in everyday settings. My solution rests on two simple but powerful ideas. First, Anchor Embedding through Feature Aggregation (AnchorEFA) creates a "super-anchor" by averaging four palmprint samples from the same person. This pooled anchor smooths out noise and highlights the consistent patterns shared between left and right palms. Second, I use a Concatenated Similarity Measurement-combining Euclidean distance with Element-wise Absolute Difference (EAD)-so the model can pick up both big structural similarities and tiny textural differences. I tested this approach on three public datasets (POLYU_Left_Right, TongjiS1_Left_Right, and CASIA_Left_Right) and saw a clear jump in accuracy compared to traditional methods. In fact, my four-sample AnchorEFA plus hybrid similarity metric did not just beat the baseline-it set a new benchmark for cross-palmprint recognition. In short, recognizing a palmprint from its opposite pair is not just feasible-it is practical, accurate, and ready for real-world use. This work opens the door to more secure, user-friendly biometric systems that still work even when only one palmprint is available.

This study addresses the pressing need for improved lung cancer diagnosis and treatment by leveraging computational methods and omics data analysis. Lung cancer remains a leading cause of cancer-related deaths globally, highlighting the urgency for more effective diagnostic and therapeutic approaches. Current diagnostic methods, such as imaging and biopsies, suffer from limitations in sensitivity, specificity, and accessibility, often due to factors such as poor data quality, small sample sizes, and variability in data sources. These limitations highlight the necessity for the development of advanced noninvasive techniques. Computational methods utilizing omics data have shown promise in overcoming these challenges by comprehensively understanding the molecular pathways involved in lung cancer. We propose a novel approach that utilizes RNA-Seq data and employs LASSO regression with attention mechanisms to identify lung cancer biomarkers. Our results demonstrate the effectiveness of this approach in identifying potential biomarkers for lung cancer, including well-known genes such as TP53, EGFR, KRAS, ALK, and PIK3CA, validating the model's ability to uncover key genes associated with lung cancer development and progression. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses revealed significant associations of the identified genes with critical biological processes and pathways, including protein synthesis, folding, cell adhesion, gene regulation, and immune responses. The PPI network analysis, constructed using the STRING database and Cytoscape application, highlighted a highly interconnected interaction landscape, with central hub genes playing pivotal roles in lung cancer progression. RPSA emerged as a crucial hub gene, consistently identified across different centrality measures. This study sheds light on the potential of computational methods and omics data analysis in improving lung cancer diagnosis and treatment, offering new insights for future research directions and personalized medicine strategies.

Identity management (IDM) systems in cloud computing struggle to securely manage user identities and access privileges in distributed environments. However, centralized IDM solutions come with high trust costs, single points of failure, and a need for appropriate security response. This paper proposes a novel decentralized IDM framework utilizing blockchain technology and automatic provisioning (AP) techniques to improve cloud computing's security, scalability, and operational efficiency. The framework employs Ethereum smart contracts and role-based access control (RBAC) to ensure secure, transparent, and automated management of user identities. Key features include support for single sign-on (SSO), multifactor authentication (MFA), and delegated proof-of-stake (DPoS) consensus for secure transaction validation. Our proposed scheme utilizes the Ethereum blockchain and smart contracts for managing user access, ensuring transparent and immutable record-keeping. The scheme introduces RBAC mechanisms to ensure precise privilege allocation and dynamic updates. The scheme also supports key IDM processes, including SSO, MFA, and lifecycle management of identities. The framework incorporates DPoS consensus to enhance security for efficient transaction validation and the prevention of fraud. To address fraudulent activities, the scheme uses machine learning to detect blockchain fraud with 99.1% accuracy, demonstrating robustness and efficiency for large-scale cloud infrastructures.

Background/Objectives: This study aims to address the critical need for accessible, early, and accurate cardiac di-agnostics, especially in resource-limited or remote settings. By shifting focus from traditional multi-lead ECG analysis to single-lead ECG data, this research explores the potential of advanced deep learning models for classifying cardiac conditions, including Nor-mal, Abnormal, Previous Myocardial Infarction (PMI), and Myocardial Infarction (MI). Methods: Five state-of-the-art deep learning architectures—Inception, DenseNet201, MobileNetV2, NASNetLarge, and VGG16—were systematically evaluated on individual ECG leads. Key performance metrics, such as model accuracy, inference time, and size, were analyzed to determine the optimal configurations for practical applications. Results: VGG16 emerged as the most accurate model, achieving an F1-score of 98.11% on lead V4 with a prediction time of 4.2 ms and a size of 528 MB, making it suitable for high-precision clinical settings. MobileNetV2, with a compact size of 13.4 MB, offered a balanced performance, achieving a 97.24% F1-score with a faster inference time of 3.2 ms, positioning it as an ideal candidate for real-time monitoring and telehealth applications. Conclusions: This study bridges a critical gap in cardiac diagnostics by demonstrating the feasibility of lightweight, scalable, single-lead ECG analysis using advanced deep learning models. The findings pave the way for deploying portable diagnostic tools across diverse settings, enhancing the accessibility and efficiency of cardiac care globally.

Ensuring safe transportation requires a comprehensive understanding of driving behaviors and road safety to mitigate traffic crashes, reduce risks and enhance mobility. This study introduces an AI-driven machine learning (ML) framework for traffic crash severity prediction, utilizing a large-scale dataset of over 2.26 million records. By integrating human, crash-specific, and vehicle-related factors, the model improves predictive accuracy and reliability. The methodology incorporates feature engineering, clustering techniques such as K-Means and HDBSCAN, with oversampling methods such as RandomOverSampler, SMOTE, Borderline-SMOTE, and ADASYN to address class imbalance, along with Correlation-Based Feature Selection (CFS) and Recursive Feature Elimination (RFE) for optimal feature selection. Among the evaluated classifiers, the Extra Trees (ET Classifier) ensemble model demonstrated superior performance, achieving 96.19% accuracy and an F1-score (macro) of 95.28%, ensuring a well-balanced prediction system. The proposed framework provides a scalable, AI-powered solution for traffic safety, offering actionable insights for intelligent transportation systems (ITS) and accident prevention strategies. By leveraging advanced ML and feature selection techniques, this approach enhances traffic risk assessment and enables data-driven decision-making.

Unnecessary days in the hospital may lead to increased hospital-acquired patient complications. So, predicting the length of stay (LOS) for patients is crucial for hospitals to optimize resource efficiency and deliver quality treatment. However, the healthcare industry faces challenges in accessing large-scale training data due to confidentiality concerns. In this paper, we apply seven machine-learning models to ten datasets collected from ten hospitals to investigate. The comparison study aims to investigate the performance of the applied models and produce the fittest model to manipulate the selected dataset and properly solve the length of stay problem. Also, we test the performance of these models using three evaluation metrics. The results show that the eXtreme Gradient Boosting (XGBoost) model outperforms the other machine learning models.

This study explores applying advanced machine-learning strategies, particularly improved semantic vectors, to predict the severity of drug-drug interactions (DDIs), a crucial element in pharmacovigilance. Based on the Adverse Event Reporting System (FAERS), our study aims to analyze the combination of advanced embedding techniques with state-of-the-art machine learning (ML) algorithms to identify and quantify DDI severity. The CatBoost Classifier is the center of our analysis, as it has emerged as the most effective model in the examined trials. We improved the performance by increasing the BioWordVec Indication Substance embedding specificity, a new creation constructed through transfer learning methodologies employed on the BioWordVec model. This approach employs not only the names of the drugs but also the indications for the drugs and the active substances, forming a highly semantic network capable of capturing multiple relations between drugs. Applying BioWordVec Indication Substance embedding combined with the CatBoost Classifier, especially using the contact-vectors method, provided the best F1 score of 73. 32% and an ROC AUC score of 84%. The results imply that this method effectively models and predicts severe consequences of DDIs using deep learning that comprehensively covers pharmacological and clinical aspects. Based on our results, we suggest incorporating semantic embedding and ML into the pharmacovigilance processes to improve the predictive potential of DDI evaluations. Thus, by enhancing the body of knowledge related to the analytical methods of assessing drug interactions, the present study substantially enhances the quality of clinical decision-making and patient protection. The novel embedding marks a significant step forward in the methodology, providing a more solid tool for the fine-grained dissection of the complexities needed in modern medicine, where multiple drug therapies are now the norm.

This paper addresses the computational methods and challenges associated with prime number generation, a critical component in encryption algorithms for ensuring data security. The generation of prime numbers efficiently is a critical challenge in various domains, including cryptography, number theory, and computer science. The quest to find more effective algorithms for prime number generation is driven by the increasing demand for secure communication and data storage and the need for efficient algorithms to solve complex mathematical problems. Our goal is to address this challenge by presenting two novel algorithms for generating prime numbers: one that generates primes up to a given limit and another that generates primes within a specified range. These innovative algorithms are founded on the formulas of odd-composed numbers, allowing them to achieve remarkable performance improvements compared to existing prime number generation algorithms. Our comprehensive experimental results reveal that our proposed algorithms outperform well-established prime number generation algorithms such as Miller-Rabin, Sieve of Atkin, Sieve of Eratosthenes, and Sieve of Sundaram regarding mean execution time. More notably, our algorithms exhibit the unique ability to provide prime numbers from range to range with a commendable performance. This substantial enhancement in performance and adaptability can significantly impact the effectiveness of various applications that depend on prime numbers, from cryptographic systems to distributed computing. By providing an efficient and flexible method for generating prime numbers, our proposed algorithms can develop more secure and reliable communication systems, enable faster computations in number theory, and support advanced computer science and mathematics research.

In this study, we introduce a novel combination of layer-static-weighted attention and ascending feature selection techniques to predict the seriousness level of adverse drug events using the Food and Drug Administration (FDA) Adverse Event Reporting System (FAERS). We utilized natural language processing (NLP) to analyze the terms in the active substance field, in addition to considering demographic and event information such as patient sex, healthcare provider qualification, and drug characterization. Our ascending feature selection method, which progressively incorporates additional features based on their importance, demonstrated continuous enhancements in prediction performance. Simultaneously, we employed a layer-static-weighted attention technique, which dynamically adjusts the model’s focus between natural language processing (NLP) and demographic features. This technique achieved its best performance at a balanced weight of 50%, yielding an average test accuracy of 74.56% and CV ROC score of 0.83 when 4000 features were included, indicating a compelling advantage to include a larger volume of meaningful features. By integrating these methodologies, we constructed a robust model capable of effectively predicting seriousness levels, offering significant potential for improving pharmacovigilance and enhancing drug safety monitoring. The results underscore the value of NLP and demographic data in predicting drug event seriousness and demonstrate the effectiveness of our combined techniques. We encourage further research to refine these methods and evaluate their application to other clinical datasets.

Heart attacks are a leading cause of mortality worldwide, necessitating the development of accurate predictive models to enhance early detection and intervention strategies. This study addresses the significant problem of class imbalance in medical datasets, specifically focusing on heart attack prediction using the Behavioral Risk Factor Surveillance System (BRFSS) dataset. To tackle this challenge, advanced machine learning (ML) methods are proposed to involve a refined dataset of 399,875 instances, with 47 significant features maintained through rigorous data cleaning and preparation. Balanced accuracy and macro-recall were chosen as primary metrics to ensure fair performance evaluation across classes in the imbalanced dataset. Our proposed system entails a detailed evaluation of various algorithms known for their effectiveness in managing class imbalance. The LGBM Classifier, XGB Classifier, and Logistic Regression (LR) are optimized using recursive feature elimination and hyperparameter tuning with Optuna. The results of this study are encapsulated in an ensemble model that significantly enhances predictive accuracy. The final model achieved 80.75% balanced accuracy and 79.97% recall for critical heart attack cases (class 1), along with an AUC score of 88.9%, indicating superior class distinction capability. Additionally, the application of SHAP (SHapley Additive exPlanations) analysis provided valuable insights into the contribution of each feature to heart attack likelihood, thus improving model transparency. This study’s successful integration of complex ML techniques with interpretability analyses like SHAP marks a substantial advance in early detection and intervention strategies in healthcare. It demonstrates the potential of sophisticated ML approaches for early heart attack detection and prevention, highlighting their value in improving outcomes for patients with chronic diseases. These findings suggest promising pathways for employing advanced analytical tools in healthcare to enhance patient care.