Research Title: Cluster preserving recursive feature elimination framework for intrusion detection in smart industry systems

Author: AHMAD MOHAMMAD SALEH KHASAWNEH, Published Year: 2026

Discover Internet of Things, 6

Faculty: Information Technology

Abstract: Cyber-physical systems, Industrial IoT, and Industry 4.0 infrastructures are converging rapidly, due to which the attack surface of industrial environments has grown substantially. Intrusion Detection Systems (IDS) are now central to operational security, yet conventional approaches still struggle, false-positive rates remain high, distinguishing genuine cyberattacks from ordinary industrial anomalies is difficult, and the computational cost of running large models often makes deployment on resource-constrained edge devices impractical. Data scarcity compounds the problem further, since labeling representative industrial attack traffic is expensive and slow. Feature selection methods such as Recursive Feature Elimination (RFE) help by reducing dimensionality, but they evaluate features one at a time and so tend to miss the higher-order interactions between features that often carry the strongest signal for identifying attacks. This paper introduces D-FOCUS-X, a Feature- Interaction-Aware Recursive Feature Elimination (FIA-RFE) framework built specifically to capture and preserve these inter-feature dependencies during selection rather than discarding them. Its core mechanism combines an interaction-aware scoring step with a cluster-preserving elimination strategy, so that groups of features which work together discriminatively are kept intact even as redundant attributes are pruned away. The result is a compact feature set that retains the information needed for accurate detection while being light enough for industrial edge deployment. On the CICIDS2017 benchmark, D-FOCUS-X reduces the original 78-dimensional feature space down to just 27 features a 65.4% reduction without sacrificing detection performance. Random Forest (RF) and XGBoost both reach 99.9% classification accuracy on this reduced set, outperforming conventional RFE, SelectKBest, and Information Gain. All results were obtained under a strictly leakage-free pipeline, and 10-fold cross-validation confirms stable generalization, with standard deviation below 0.003 across all classifiers. Together, these findings position D-FOCUS-X as a practical and computationally efficient feature selection approach for next-generation industrial IDS.

Keywords: Intrusion detection system, Industry 4.0, Cybersecurity, Feature engineering, Machine learning, Recursive feature elimination, Feature optimization

Research Title: The Effect of Digital Psychoeducation and ICU Delirium Care Bundle Programmes on Nurses' Competency, Confidence and Care-Related Strain: A Systematic Review and Narrative Synthesis

Author: Rabia Hani Amin Haddad, Published Year: 2026

Nursing in Critical Care , Volume31, Issue5

Faculty: Nursing

Abstract: Background Delirium is a common and serious complication among critically ill patients, affecting approximately 20%–80% of ICU populations and contributing to increased mortality, prolonged mechanical ventilation, extended hospital stays, cognitive impairment and higher healthcare costs. Nurses play a central role in delirium prevention, assessment and management; however, deficiencies in delirium-related knowledge, confidence and assessment practices remain widespread. Digital psychoeducation and delirium care bundle interventions have been increasingly implemented to address these gaps. Aim To systematically review and narratively synthesize evidence regarding the effects of digital psychoeducation and ICU delirium care bundle programmes on nurses' competency, confidence and care-related strain. Study Design A systematic review with narrative synthesis was conducted in accordance with PRISMA 2020 and SWiM guidance. Six electronic databases (PubMed, CINAHL, Scopus, Web of Science, Embase and ProQuest) were searched. Eligible studies included quasi-experimental, pre–post and quality-improvement studies evaluating delirium-focused educational or care bundle interventions among ICU nurses. Methodological quality was assessed using the Joanna Briggs Institute Critical Appraisal Checklist for Quasi-Experimental Studies. Results Eight studies involving approximately 472 nurses from five countries met the inclusion criteria. Competency outcomes improved in 87.5% of studies, with delirium knowledge increasing by 100%–150%, CAM-ICU assessment accuracy reaching 95.5%, documentation completeness exceeding 91% and bundle compliance reaching 97.5%. Confidence improved across all studies evaluating this outcome, with increases ranging from 35.3 to 60.1 percentage points, including one study reporting a very large effect size (Cohen's d = 2.31). The only study assessing care-related strain demonstrated a 48.8% reduction in delirium-related nursing stress. Overall, educational interventions were associated with improved clinical practice indicators, although implementation varied across healthcare settings. Conclusion Digital psychoeducation and ICU delirium care bundle programmes consistently improve nurses' delirium-related competency and confidence and may reduce care-related strain. However, evidence regarding strain remains limited, and variability in clinical implementation highlights the need for more rigorous controlled studies and long-term evaluations to strengthen the evidence base. Relevance to Clinical Practice Integrating digital delirium education programmes and structured care bundles into ICU nursing practice can improve nurses' competency, confidence and adherence to evidence-based delirium assessment, supporting earlier recognition and higher quality patient care. Trial Registration CRD4201380975

Keywords: CAM-ICU | care-related strain | competency | confidence | delirium | delirium care bundle | ICU nurses | intensive care unit |narrative synthesis | psychoeducation (PDF) The Effect of Digital Psychoeducation and ICU Delirium Care Bundle Programmes on Nurses' Competency, Confidence and Care‐Related Strain: A Systematic Review and Narrative Synthesis. Available from: https://www.researchgate.net/publication/413595227_The_Effect_of_Digital_Psychoeducation_and_ICU_Delirium_Care_Bundle_Programmes_on_Nurses'_Competency_Confidence_and_Care-Related_Strain_A_Systematic_Review_and_Narrative_Synthesis [accessed Oct 15 2026].

Research Title: TrustEdge-V2X: Deployment-Aware Edge Intelligence for V2X/IoV Intrusion and Misbehavior Detection

Author: Mina Samir Shenouda Abd elnour, Published Year: 2026

Computers, 15

Faculty: Information Technology

Abstract: Most studies on vehicular cybersecurity focus on classification accuracy but fail to evaluate the applicability of the models to the edge. This paper proposes a secure edge intelligence framework for V2X/IoV intrusion and misbehavior detection, called TrustEdge-V2X, which is deployment-aware. The framework combines dataset-role qualification, attack taxonomy, AI model benchmarking, feature-budget analysis, deployment ranking based on EdgeScore, offline risk-aware orchestration, external validation, robustness testing, explainable AI, repeated-run statistical analysis and ablation studies. Three datasets are assigned different experimental roles: VeReMi_NextGen is used for core V2X/VANET misbehavior detection, CICIoT2023 is used for supporting edge/IoT intrusion experiments and HCRL_CarHacking is used for external IoV/CAN validation. LightGBM outperformed all other AI models in EdgeScore (0.9383), F1-score (0.9878), MCC (0.9758), and inference latency (0.009419 ms per sample) across all eight AI models and six scenarios on VeReMi_NextGen for binary detection. In five dataset-task cases, the accuracy-best model was different from the EdgeScore-best model, which is the most important point to note: the best model in terms of accuracy is not necessarily the best model in terms of EdgeScore. Compact feature subsets were competitive, and robustness testing demonstrated an average F1 decrease of 0.1423 when tested under stress. The orchestration layer was found to be beneficial for the tasks, but it did not always perform better than the best fixed policy. As a whole, TrustEdge-V2X offers a systematic approach to the assessment and selection of vehicular cybersecurity models based on the operational and deployment conditions, not only on the classification accuracy.

Keywords: V2X; IoV; VANET security; intrusion detection; misbehavior detection; edge intelligence; Edge AI; EdgeScore; orchestration; robustness; XAI

Research Title: A Preprocessing and Analytics Statistical Approach Using Demographic Data on Alzheimer’s Disease Diagnosis

Author: Ali Ahmad Alawneh, Published Year: 2026

Journal of Information and Knowledge Management, 4

Faculty: Business

Abstract: Abstract. Alzheimer’s disease (AD) is one of the leading causes of brain degeneration, memory impairment and physical functionality of elderly people around the world. In addition, this disease might impact patients’ family members and the financial, economic and social aspects of their societies. Such a prevailing disease necessitates the diagnosis and prognosis of its inception, development and progression as early as possible. Researchers have recently investigated different statistical, data analytics and machine learning approaches to detect such disease at an earlier stage in order to help neurologists to detect the disease earlier, before its progression and with minimal harm. This paper reports the empirical study employing data analytics and statistics performed on the Alzheimer’s Disease Neuroimaging Initiative longitudinal data repository (ADNI) to assess the impact of demographic factors such as gender, age, education, race, ethnicity and marital status on AD progression. These factors often influence AD through its progression from cognitively normal (CN) status to mild cognitive impairment (MCI) to AD. The study utilises statistical techniques including descriptive analytics, cross-tabulation distributions, Chi-square tests, ANOVA analyses and box plot visualisations on the ADNI dataset to assess the effect of demographic factors on AD diagnosis. Statistical analysis results reveal significant relationships between demographic features (age, gender, education, ethnicity, race and marital status) and the progression of the disease through the three clinical states (CN, MCI and Dementia) that can significantly assist in the determination of early AD.

Keywords: Keywords: Alzheimer’s disease; demographic data; data preprocessing; data analytics; ADNI.

Research Title: Explainable and Deployment-Aware Zero-Day Intrusion Detection for Cloud-Level Backend and Management Ecosystems in EV/V2X Cyber–Physical Systems

Author: AHMED ABDELRAHIM ALI EIDOUH, Published Year: 2026

Computers, 15

Faculty: Information Technology

Abstract: Abstract With the escalating frequency of sophisticated zero-day attacks, overcoming the critical limitations of signature-based Intrusion Detection Systems (IDSs) has become paramount. This study proposes a hybrid multi-layered intrusion detection framework combining traditional machine learning, Deep Neural Architectures (DenseNN), and ensemble methods to evaluate zero-day resilience within cloud-level backend connectivity interfacing EV and V2X management ecosystems. Using the comprehensive CSE-CIC-IDS2018 benchmark as a surrogate environment, a code-executed Leave-One-Attack-Out (LOAO) cross-validation protocol across 13 distinct attack families was implemented to assess unseen-attack-family generalization within the benchmark to unseen threats. Furthermore, Explainable Artificial Intelligence (XAI) auditing, utilizing SHapley Additive exPlanations (SHAP) and Integrated Gradients, was integrated to inspect decision boundaries and resolve feature-attribution failure modes. Critically, the audit identified an artifact-driven data leakage caused by the Timestamp and identifier features, demonstrating that models learned temporal schedules rather than behavioral network signatures. Re-executing all experiments post-leakage removal quantified performance drops across all classifiers (e.g., Gaussian NB dropping by up to 20.88 percentage points in accuracy (at the 60% training ratio; 18.30 points at the 80% ratio)). Under standard binary classification metrics, tree ensembles (Random Forest and Extra Trees) achieved high in-distribution detection (F1 > 0.95) with rapid inference latency (≈0.05–−0.07 ms/sample). However, the rigorous LOAO evaluation revealed a substantial generalization penalty on truly unseen zero-day families (e.g., SQL Injection and Infiltration), where simpler linear models demonstrated broader generalization robustness (mean LOAO F1 = 0.397) compared with complex tree-ensemble models. By rectifying dataset leakage and benchmarking deployment trade-offs (training runtime, throughput, and memory footprint), this study delivers actionable, transparent guidelines for deployment-oriented IDS evaluation in dynamic network infrastructures.

Keywords: zero-day attacks; intrusion detection systems; Explainable AI (XAI); machine learning; CSE-CIC-IDS2018; data leakage; feature importance

Research Title: An Intelligent Computational Framework for Multi-Criteria Decision Analysis under Uncertainty

Author: AHMED ABDELRAHIM ALI EIDOUH, Published Year: 2026

Statistics, Optimization & Information Computing, 16

Faculty: Information Technology

Abstract: Uncertainty, vagueness, and inadequate knowledge sometimes make it hard to make decisions in complicated real-world situations. This research provides an intelligent computational paradigm for multi-criteria decision analysis under uncertainty to address these problems. The suggested approach combines rough set theory with advanced methods for modelling uncertainty to accurately show decision information that is unclear or contradictory. A structured decision-support method is created to assess and prioritize many choices according to a range of criteria that may be conflicting. A sustainable system selection problem, which takes into account technical, environmental, and economic factors all at once, shows how the proposed framework can be used. The experimental findings demonstrate that the suggested methodology improves decision robustness, interpretability, and reliability in comparison to current decision-making models. The suggested intelligent computational framework can be adapted for various decision-support and artificial intelligence applications functioning in uncertain environments, due to its flexibility and universality.

Keywords: Intelligent Decision Systems, Computational Decision-Making, Multi-Criteria Decision Analysis, Uncertainty Modelling

Research Title: Adaptive Differential Privacy in Federated Learning for Privacy-Preserving ICU Clinical Decision Support at the Edge

Author: Ahmad Suleman Qasim Alnawasrah, Published Year: 2026

International Journal of Advanced Computer Science and Applications, 17

Faculty: Information Technology

Abstract: Clinical decision assistance that protects patient privacy is a prominent research issue in federated learning, edge computing, and intensive care analytics. However, typical differential privacy algorithms allocate equal privacy budgets to all clinical parameters regardless of sensitivity and predictive power. This may lower privacy-utility trade-offs. This study develops an adaptive differential privacy mechanism for federated learning that dynamically adjusts per-feature Laplace noise and privacy budgets to improve privacy protection while preserving clinical prediction performance. Adaptive Differential Privacy in Federated Learning (ADP-FL) dynamically adjusts per-feature Laplace noise and privacy budgets based on feature-scored sensitivity considering clinical relevance, statistical impact, and re-identification risk. The method was evaluated on the eICU Collaborative Research Database containing 113,265 ICU records and compared against Uniform Differential Privacy and TimeAdaptive Differential Privacy across different privacy budgets (ε). Experiments conducted on the eICU Collaborative Research Database achieved test accuracies of up to 82.7%. Across the evaluated privacy budgets, Feature-Adaptive Differential Privacy consistently outperformed both Uniform and Time-Adaptive Differential Privacy, achieving an F1-score of 0.6672 at ε = 0.5 while maintaining performance close to the no-DP baseline. These findings demonstrate that sensitivity-aware feature-level privacy budgeting preserves key predictive dimensions while minimizing unnecessary perturbation of less sensitive features, providing a robust foundation for privacy-compliant edge-based clinical decision support systems. The novelty of this study lies in its feature-scored adaptive differential privacy mechanism, which dynamically allocates per-feature privacy budgets and Laplace noise in federated learning for privacy-preserving edge-based clinical decision support.

Keywords: Federated Learning; Differential Privacy; Clinical Decision Support; Edge Computing; ICU Data.

Research Title: Effect Of Inspiratory Muscle Training Combined with Low-Level Light Therapy on Pulmonary Function and Pain on Neck Pain Post Thyroidectomy Scar Wound in Premenopausal Females

Author: Mohamed Ali Elsayed Abdullah, Published Year: 2026

Natural Resources for Human Health, 6

Faculty: Allied Medical Sciences

Abstract: Background: About 80% of patients experience neck pain and stiffness after having an open thyroidectomy, which is one of the most common endocrine surgical procedures. This procedure is typically done as a result of thyroid cancer. Purpose: The goal of the study was to examine the efficacy of a combination of inspiratory muscle training along with low-level light therapy for post-thyroidectomy neck pain among premenopausal women in terms of pulmonary function and pain, as well as wound healing. Methods: 60 female patients diagnosed to get open surgery of thyroidectomy, their ages between 35 and 45 years, and their Body Mass Index (BMI) ranged between 25 and 29.9 kg/m². They were randomized into three groups. Group (A) consisted of 20 female patients who received only traditional medical treatment daily at first week, then three sessions per week for an additional three weeks, while Group (B) consisted of 20 female patients who received traditional medical treatment and inspiratory muscle training daily at first week, then three sessions per week for an additional three weeks. Group (C) consisted of 20 female patients who received traditional medical treatment, inspiratory muscle training, and low-level light therapy daily for the first week, then three sessions per week, day after day, for an additional three weeks. FVC, FEV1, PEF, and MVV, as well as PImax during the spirometry test, visual analog scale, and photographic wound assessment, were assessed pre- and post-study. Results: Group A showed non-significant improvement pre our post-treatment in all outcome measurements, but both groups A and B showed significant improvement in all outcome measurements (pulmonary function, pain (VAS score), and wound healing) from pre-treatment to post-treatment measurements (P value < 0.001). Conclusion: Both inspiratory muscle training alone and inspiratory muscle training combined with low-level light therapy have a beneficial effect on improving pulmonary function, pain (VAS score), and wound healing in favor of the combination of inspiratory muscle training with low-level light therapy.

Keywords: inspiratory muscle training, low-level light therapy, pulmonary function, neck pain, post-thyroidectomy, premenopausal females.

Research Title: Adaptive Differential Privacy in Federated Learning for Privacy-Preserving ICU Clinical Decision Support at the Edge

Author: Ahmad Suleman Qasim Alnawasrah, Published Year: 2026

International Journal of Advanced Computer Science and Applications, 17

Faculty: Information Technology

Abstract: Clinical decision assistance that protects patient privacy is a prominent research issue in federated learning, edge computing, and intensive care analytics. However, typical differential privacy algorithms allocate equal privacy budgets to all clinical parameters regardless of sensitivity and predictive power. This may lower privacy-utility trade-offs. This study develops an adaptive differential privacy mechanism for federated learning that dynamically adjusts per-feature Laplace noise and privacy budgets to improve privacy protection while preserving clinical prediction performance. Adaptive Differential Privacy in Federated Learning (ADP-FL) dynamically adjusts per-feature Laplace noise and privacy budgets based on feature-scored sensitivity considering clinical relevance, statistical impact, and re-identification risk. The method was evaluated on the eICU Collaborative Research Database containing 113,265 ICU records and compared against Uniform Differential Privacy and Time Adaptive Differential Privacy across different privacy budgets (ε). Experiments conducted on the eICU Collaborative Research Database achieved test accuracies of up to 82.7%. Across the evaluated privacy budgets, Feature-Adaptive Differential Privacy consistently outperformed both Uniform and Time-Adaptive Dif ferential Privacy, achieving an F1-score of 0.6672 at ε = 0.5 while maintaining performance close to the no-DP baseline. These findings demonstrate that sensitivity-aware feature-level privacy budgeting preserves key predictive dimensions while minimizing unnecessary perturbation of less sensitive features, providing a robust foundation for privacy-compliant edge-based clinical decision support systems. The novelty of this study lies in its feature-scored adaptive differential privacy mechanism, which dynamically allocates per-feature privacy budgets and Laplace noise in federated learning for privacy-preserving edge-based clinical decision support.

Keywords: Federated learning; differential privacy; clinical decision support; edge computing; ICU data

Research Title: The influence of website quality, e-WOM, and e-trust on consumers’ purchase intention

Author: Mohammad Assad Mohammad Hasan, Published Year: 2026

International Journal of Data and Network Science, 10

Faculty: Business

Abstract: The increasingly competitive environment has created challenging marketing conditions for Jordanian real estate companies. This research investigated the effect of website quality (WSQ) and electronic word-of-mouth (e-WOM) on customer purchase intention in Jordanian real estate companies, with e-Trust as a mediator. The research framework was guided by the Information Systems (IS) Success Model, Social Influence Theory, and the Trust Model. Questionnaires were distributed through Google Forms to respondents selected using purposive sampling, resulting in 310 valid responses. The data were analysed, and the hypotheses were tested using the PLS-SEM method. The findings reveal the significant impact of e-WOM and e-Trust on purchase intention, while website quality did not have a direct effect. e-Trust mediates the relationship between website quality, e-WOM and purchase intention. Theoretically, this study fills gaps in the literature by integrating the IS Success Model, Social Influence Theory and the Trust Model. Practically, the study shows that real estate companies can enhance their online presence and marketing strategies by prioritising e-Trust and leveraging website quality and e-WOM to effectively drive purchase intention.

Keywords: Website Quality, e-WOM, Purchase Intention, e-Trust, Real Estate, Jordan