Almaarefa University
The authoritative source for research information and outputs is Elsevier Scopus, with researcher profile verification from the Deanship of Scientific Research.

المخرجات

Research outputs

Articles, reviews, conference papers and book chapters, indexed by DOI and linked to University researchers.

ArticleOpen access202549citation· Semantic Scholar

Neonatal nurses’ experiences with generative AI in clinical decision-making: a qualitative exploration in high-risk nicus

BMC Nursing · vol. 24 (1) · 386-386

BACKGROUND: Neonatal nurses in high-risk Neonatal Intensive Care Units (NICUs) navigate complex, time-sensitive clinical decisions where accuracy and judgment are critical. Generative artificial intelligence (AI) has emerged as a supportive tool, yet its integration raises concerns about its impact on nurses' decision-making, professional autonomy, and organizational workflows. AIM: This study explored how neonatal nurses experience and integrate generative AI in clinical decision-making, examining its influence on nursing practice, organizational dynamics, and cultural adaptation in Saudi Arabian NICUs. METHODS: An interpretive phenomenological approach, guided by Complexity Science, Normalization Process Theory, and Tanner's Clinical Judgment Model, was employed. A purposive sample of 33 neonatal nurses participated in semi-structured interviews and focus groups. Thematic analysis was used to code and interpret data, supported by an inter-rater reliability of 0.88. Simple frequency counts were included to illustrate the prevalence of themes but were not used as quantitative measures. Trustworthiness was ensured through reflexive journaling, peer debriefing, and member checking. RESULTS: Five themes emerged: (1) Clinical Decision-Making, where 93.9% of nurses reported that AI-enhanced judgment but required human validation; (2) Professional Practice Transformation, with 84.8% noting evolving role boundaries and workflow changes; (3) Organizational Factors, as 97.0% emphasized the necessity of infrastructure, training, and policy integration; (4) Cultural Influences, with 87.9% highlighting AI's alignment with family-centered care; and (5) Implementation Challenges, where 90.9% identified technical barriers and adaptation strategies. CONCLUSIONS: Generative AI can support neonatal nurses in clinical decision-making, but its effectiveness depends on structured training, reliable infrastructure, and culturally sensitive implementation. These findings provide evidence-based insights for policymakers and healthcare leaders to ensure AI integration enhances nursing expertise while maintaining safe, patient-centered care.

generative AIneonatal nursingclinical decision-makinghigh-risk infants
ArticleOpen access202548citation· Semantic Scholar

Early detection of mental health disorders using machine learning models using behavioral and voice data analysis

Scientific Reports · vol. 15 (1) · 16518-16518

People of all demographics are impacted by mental illness, which has become a widespread and international health problem. Effective treatment and support for mental illnesses depend on early discovery and precise diagnosis. Notably, delayed diagnosis may lead to suicidal thoughts, destructive behaviour, and death. Manual diagnosis is time-consuming and laborious. With the advent of AI, this research aims to develop a novel mental health disorder detection network with the objective of maximum accuracy and early discovery. For this reason, this study presents a novel framework for the early detection of mental illness disorders using a multi-modal approach combining speech and behavioral data. This framework preprocesses and analyzes two distinct datasets to handle missing values, normalize data, and eliminate outliers. The proposed NeuroVibeNet combines Improved Random Forest (IRF) and Light Gradient-Boosting Machine (LightGBM) for behavioral data and Hybrid Support Vector Machine (SVM) and K-Nearest Neighbors (KNN) for voice data. Finally, a weighted voting mechanism is applied to consolidate predictions. The proposed model achieves robust performance and a competitive accuracy of 99.06% in distinguishing normal and pathological conditions. This framework validates the feasibility of multi-modal data integration for reliable and early mental illness detection.

mental disorder detectionearly diagnosisspeech analysisbehavioral data
ArticleOpen access202545citation· Semantic Scholar

Unveiling diabetes onset: Optimized XGBoost with Bayesian optimization for enhanced prediction

PLoS ONE · vol. 20 (1) · e0310218-e0310218

Diabetes, a chronic condition affecting millions worldwide, necessitates early intervention to prevent severe complications. While accurately predicting diabetes onset or progression remains challenging due to complex and imbalanced datasets, recent advancements in machine learning offer potential solutions. Traditional prediction models, often limited by default parameters, have been superseded by more sophisticated approaches. Leveraging Bayesian optimization to fine-tune XGBoost, researchers can harness the power of complex data analysis to improve predictive accuracy. By identifying key factors influencing diabetes risk, personalized prevention strategies can be developed, ultimately enhancing patient outcomes. Successful implementation requires meticulous data management, stringent ethical considerations, and seamless integration into healthcare systems. This study focused on optimizing the hyperparameters of an XGBoost ensemble machine learning model using Bayesian optimization. Compared to grid search XGBoost (accuracy: 97.24%, F1-score: 95.72%, MCC: 81.02%), the XGBoost with Bayesian optimization achieved slightly improved performance (accuracy: 97.26%, F1-score: 95.72%, MCC:81.18%). Although the improvements observed in this study are modest, the optimized XGBoost model with Bayesian optimization represents a promising step towards revolutionizing diabetes prevention and treatment. This approach holds significant potential to improve outcomes for individuals at risk of developing diabetes.

diabetes onset predictionXGBoostBayesian optimizationhyperparameter optimization
ArticleClosed202542citation· Semantic Scholar

Satellite IoT Terminal Authentication Mechanism for Enhancing Consumer Security Over Cyber–Physical Systems

IEEE Transactions on Consumer Electronics · vol. 71 (2) · 4771-4779

The satellite Internet of Things (IoT), utilizing Cyber-Physical Systems (CPS) Communication Networks and various satellite-borne and ground-based terminal devices, enables integrated human-object connectivity and information exchange across space, sky, and ground. A terminal authentication method based on trust evaluation is proposed to ensure the secure and efficient utilization of satellite IoT resources. This method optimizes authentication strategies by evaluating the trustworthiness of terminal devices. A satellite IoT terminal trust evaluation mechanism is proposed for scenarios where a previously strongly authenticated terminal needs re-authentication within a short time or where trust is granted via a strongly authenticated node. This mechanism describes dynamically adjusted trust metrics and evaluation models based on direct and indirect trust. A satellite IoT access authentication protocol based on trust evaluation is also designed, employing corresponding authentication mechanisms according to the device’s trust level. This simplifies the re-authentication process for highly trusted devices, enabling differentiated authentication services for various terminal devices within the same authentication framework. A comprehensive security analysis of the protocol is conducted using informal methods and the Tamarin formal verification tool. Experiments validate the protocol’s effectiveness, showing that it can reduce overall authentication costs and streamline the authentication process through trust-based simplifications.

satellite Internet of Thingsterminal authenticationtrust evaluationre-authentication
ArticleOpen access202540citation· Semantic Scholar

Barriers to reporting workplace violence: a qualitative study of nurses’ perceptions in tertiary care settings

BMC Nursing · vol. 24 (1) · 395-395

BACKGROUND: Workplace violence (WPV) remains a formidable concern among nurses worldwide, with up to 60% in Saudi Arabia reportedly experiencing some form of aggression. In tertiary care hospitals, robust hierarchies and cultural norms intensify underreporting, thwarting evidence-based prevention and obscuring vital data. AIM: This qualitative study investigated the perceived barriers to WPV reporting among nurses in tertiary care settings in the Aljouf region of Saudi Arabia, specifically addressing how organizational and cultural factors converge to discourage formal incident reporting. METHODS: A qualitative descriptive design was employed, guided by Ajzen's Theory of Planned Behavior and the Social Ecological Model. Thirty-six registered nurses, purposively sampled from three tertiary hospitals, participated in six semi-structured focus groups conducted in Arabic or English, depending on participant preference. Data were thematically analyzed in NVivo, with methodological rigor ensured through triangulation and inter-coder reliability. RESULTS: Three principal themes emerged: (1) Emotional and Psychological Barriers (78%), encompassing distress, anxiety, and fears of professional blame; (2) Organizational Ineffectiveness (65%), marked by convoluted reporting processes and perceived managerial indifference; and (3) Cultural and Hierarchical Influences (57%), reflecting deference to authority and normalization of violence. These themes illustrate how attitudes, subjective norms, and perceived behavioral control shaped by socio-cultural dynamics collectively contribute to persistent underreporting. CONCLUSIONS: Mitigating WPV underreporting in Saudi tertiary care hospitals requires streamlined, user-friendly reporting channels, leadership accountability, and holistic psychosocial support. Implementing interprofessional education aimed at dismantling hierarchical imbalances can foster a zero-tolerance ethos toward violence. Longitudinal and comparative research should further examine evolving reporting behaviors to refine context-specific, culturally attuned strategies for addressing WPV. CLINICAL TRIAL NUMBER: Not applicable.

workplace violenceunderreportingnursesbarriers to reporting
ArticleOpen access202533citation· Semantic Scholar

GraphFedAI framework for DDoS attack detection in IoT systems using federated learning and graph based artificial intelligence

Scientific Reports · vol. 15 (1) · 28050-28050

The Internet of Things (IoT) consists of physical objects and devices embedded with network connectivity, software, and sensors to collect and transmit data. The development of the Internet of Things (IoT) has led to various security and privacy issues, including distributed denial-of-service (DDoS) attacks. Conventional attack detection methods face significant challenges related to privacy, scalability, and adaptability due to the dynamic nature of IoT environments. To address these limitations, this research proposes GraphFedAI, a novel framework that integrates adaptive session-based graph modeling, Pearson correlation-guided feature selection, interpolation-aware graph neural network (GNN) training, and federated learning to enable robust, scalable, and privacy-preserving DDoS detection in heterogeneous Internet of Things (IoT) networks.The framework represents the IoT network as dynamic graphs where communication patterns among devices are modeled as edges that evolve over time. Graph neural networks are utilized to extract both temporal and structural features from these graphs, thereby enhancing the accuracy of DDoS detection. Federated learning is incorporated to maintain data privacy by training models locally on each device without sharing raw data. This integration also ensures system scalability, as FL adapts training based on localized network topology.The system is evaluated using the CIC-IoT-2023 dataset, demonstrating its effectiveness in achieving high detection accuracy, low false positive rates, and strong resilience under dynamic IoT conditions.

DDoS attack detectionInternet of Things securityfederated learninggraph neural networks
ArticleOpen access202533citation· Semantic Scholar

Virtual reality simulation for high-risk neonatal emergency nursing training: a mixed-methods study on nurse competency and outcomes

BMC Nursing · vol. 24 (1) · 367-367

BACKGROUND: Nurses in neonatal intensive care units face critical challenges in managing emergencies where timely, accurate interventions are essential for survival. Traditional nursing education often lacks the hands-on, immersive training required to build complex emergency skills, contributing to persistent neonatal mortality globally. Virtual reality (VR) simulation, grounded in Kolb's Experiential Learning Theory, offers a promising solution by replicating realistic and repeatable clinical scenarios. While VR has shown potential in nursing education, its specific impact on high-risk neonatal emergencies remains underexplored. AIM: To evaluate the effectiveness of a VR simulation program in enhancing nurse competency and improving neonatal outcomes during emergency care, grounded in Kolb's Experiential Learning Theory. METHODS: A concurrent triangulation mixed-methods design was implemented over two weeks across four pediatric hospitals. Through stratified random sampling, 128 NICU nurses were allocated to VR simulation (n = 64) or traditional training (n = 64) groups. Quantitative data were collected using validated instruments (OSCE: CVI = 0.92, MCQ: α = 0.86) measuring clinical skills, knowledge retention, and decision-making accuracy. Qualitative data were gathered through semi-structured interviews (n = 24) exploring experiential aspects. RESULTS: The VR group showed significant improvements in clinical skills (OSCE: +16.1 points, p < 0.001, d = 1.58), decision-making accuracy (+ 16.7%, p < 0.001), and reduced stabilization times (-6.2 min, p < 0.001). Patient safety events decreased by 52% (p < 0.001). Thematic analysis revealed enhanced professional competence (83%), reduced clinical anxiety (75%), and positive learning experiences (88%), despite minor technical challenges. CONCLUSION: VR simulation demonstrates superior effectiveness for neonatal emergency training, significantly improving both nurse competency and patient outcomes. While geographic specificity and brief follow-up duration limit generalizability, findings support VR's potential for enhancing emergency preparedness. Future research should address longitudinal outcomes and implementation across diverse healthcare settings. CLINICAL TRIAL NUMBER: Not applicable.

virtual reality simulationneonatal intensive care unitnursing competenciesKolb's experiential learning theory
ArticleOpen access202533citation· Semantic Scholar

Opportunistic access control scheme for enhancing IoT-enabled healthcare security using blockchain and machine learning

Scientific Reports · vol. 15 (1) · 7589-7589

The healthcare industry, aided by technology, leverages the Internet of Things (IoT) paradigm to offer patient/user-related services that are ubiquitous and personalized. The authorized repository stores ubiquitous data for which access-level securities are granted. These security measures ensure that only authorized entities can access patient/user health information, preventing unauthorized entries and data downloads. However, recent sophisticated security and privacy attacks such as data breaches, data integrity issues, and data collusion have raised concerns in the healthcare industry. As healthcare data grows, conventional solutions often fail due to scalability concerns, causing inefficiencies and delays. This is especially true for multi-key authentication. Dependence on conventional access control systems leads to security flaws and authorization errors caused by static user behaviour models. This article introduces an Opportunistic Access Control Scheme (OACS) for leveraging access-level security. This approach is a defendable access control scheme in which the user permissions are based on their requirement and data. After accessing the healthcare record, a centralized IoT security augmentation and assessment is provided. The blockchain records determine and revoke the access grant based on previous access and delegation sequences. This scheme analyses the possible delegation methods for providing precise users with interrupt-free healthcare record access. The blockchain recommendations are analyzed using a trained learning paradigm to provide further access and denials. The proposed method reduces false rates by 11.74%, increases access rates by 13.1%, speeds up access and processing by 12.36% and 13.23%, respectively, and reduces failure rates by 9.94%. The OACS decreases false rates by 10.64%, processing time by 15.62%, and failure rates by 10.95%.

Internet of Thingshealthcare data securityblockchainmachine learning
ArticleOpen access202533citation· Semantic Scholar

Multi-Stage Neural Network-Based Ensemble Learning Approach for Wheat Leaf Disease Classification

IEEE Access · vol. 13 · 30101-30116

Prompt and accurate classification of wheat leaf diseases and their severity is essential for precise diagnosis, effective pesticide application, efficient disease control, and enhanced wheat production and quality. Nevertheless, the wide range of wheat diseases poses a significant challenge in terms of their identification, especially in intricate agricultural landscapes. The utilization of conventional models has major limitations in wheat disease detection, including dataset-specific performance, overfitting due to limited data, and high computational needs, making deployment in resource-constrained situations difficult. To address these challenges, this research proposes a multi-stage Convolutional Neural Network (CNN) based Ensemble Learning (EL) approach, which utilizes several concurrent CNN models with a bagged EL method to classify wheat leaf diseases. The various stages in the EL approach employ a multi-level framework to enhance feature extraction and capture complex data patterns. This improves the model’s emphasis on diseases while reducing the influence of complex backgrounds on disease identification. The proposed method, integrating pretrained CNNs and a bagging ensemble technique, achieved an accuracy of 86.78%, 98.28%, and 99.16% on the three publicly available datasets, outperforming the state-of-the-art models. These results demonstrate the potential of the proposed model for real-time disease diagnosis.

ensemble learningbaggingdisease severity assessmentplant disease diagnosis
ArticleOpen access202532citation· Semantic Scholar

Multi-scale feature fusion of deep convolutional neural networks on cancerous tumor detection and classification using biomedical images

Scientific Reports · vol. 15 (1) · 1105-1105

In the present scenario, cancerous tumours are common in humans due to major changes in nearby environments. Skin cancer is a considerable disease detected among people. This cancer is the uncontrolled evolution of atypical skin cells. It occurs when DNA injury to skin cells, or a genetic defect, leads to an increase quickly and establishes malignant tumors. However, in rare instances, many types of skin cancer occur from DNA changes tempted by infrared light affecting skin cells. This disease is a worldwide health problem, so an accurate and appropriate diagnosis is needed for efficient treatment. Current developments in medical technology, like smart recognition and analysis utilizing machine learning (ML) and deep learning (DL) techniques, have transformed the analysis and treatment of these conditions. These approaches will be highly effective for the recognition of skin cancer utilizing biomedical imaging. This study develops a Multi-scale Feature Fusion of Deep Convolutional Neural Networks on Cancerous Tumor Detection and Classification (MFFDCNN-CTDC) model. The main aim of the MFFDCNN-CTDC model is to detect and classify cancerous tumours using biomedical imaging. To eliminate unwanted noise, the MFFDCNN-CTDC method initially utilizes a sobel filter (SF) for the image preprocessing stage. For the segmentation process, Unet3+ is employed, providing precise localization of tumour regions. Next, the MFFDCNN-CTDC model incorporates multi-scale feature fusion by combining ResNet50 and EfficientNet architectures, capitalizing on their complementary strengths in feature extraction from varying depths and scales of the input images. The convolutional autoencoder (CAE) model is utilized for the classification method. Finally, the parameter tuning process is performed through a hybrid fireworks whale optimization algorithm (FWWOA) to enhance the classification performance of the CAE model. A wide range of experiments is performed to authorize the performance of the MFFDCNN-CTDC approach. The experimental validation of the MFFDCNN-CTDC approach exhibited a superior accuracy value of 98.78% and 99.02% over existing techniques under ISIC 2017 and HAM10000 datasets.

skin cancer detectionskin lesion classificationmulti-scale feature fusionconvolutional neural networks
ArticleOpen access202529citation· Semantic Scholar

Optimizing IoT intrusion detection with cosine similarity based dataset balancing and hybrid deep learning

Scientific Reports · vol. 15 (1) · 30939-30939

With IoT networks expected to exceed 29 billion connected devices by 2030, the risk of cyberattacks has never been higher. As more devices come online, the attack surface for hackers continues to expand, making cybersecurity a pressing concern. Intrusion Detection Systems (IDS) are essential for identifying and mitigating these threats in real-time. However, a significant challenge IDS faces is dealing with imbalanced datasets, where attack instances are significantly underrepresented compared to normal traffic. Training models on such skewed data leads to a bias toward majority-class patterns, reducing their ability to detect intrusions effectively. To address this issue, this work introduces CSMCR (Cosine Similarity-based Majority Class Reduction), a novel technique that selectively removes redundant majority-class samples while preserving dataset integrity. Unlike traditional approaches like SMOTE (oversampling) or random undersampling, CSMCR ensures that the retained majority instances remain diverse by analyzing feature-wise similarity. This prevents unnecessary data duplication and minimizes information loss. Additionally, we developed a hybrid deep learning model integrating RegNet and FBNet architectures to enhance feature extraction and classification performance. Experimental results on multiple IDS datasets confirm that balancing the dataset to a 1:1 ratio optimally prevents overfitting and improves model interpretability. The proposed model achieved an F1-score of 0.9758 on RT-IoT2022 and 0.9275 on UNSW Bot-IoT, outperforming SMOTE-based methods in accuracy and computational efficiency. Notably, CSMCR reduced training time by 53% compared to conventional oversampling techniques. Incremental training evaluations reveal that bias formation reduces performance beyond a 1:2 majority-to-minority ratio. These findings establish CSMCR as a robust, scalable, and computationally efficient IDS balancing strategy tailored for IoT network security.

IoT intrusion detectionclass imbalancecosine similaritydata balancing
ArticleClosed202529citation· Semantic Scholar

Perceptions of effectiveness and ethical use of AI tools in academic writing: A study Among PhD scholars in India

Information Development · vol. 41 (3) · 728-746

This study explores the perceptions, effectiveness, and ethical implications of AI tools in academic writing among PhD scholars in India. It evaluates the comparative effectiveness of AI tools versus traditional methods for paraphrasing, plagiarism detection, and citation improvement. The research also investigates user confidence in AI tools’ adaptability to various academic disciplines and their potential to support critical thinking and academic integrity. A structured quantitative approach was employed, utilizing validated surveys with 184 respondents from the University of Mumbai. Findings indicate that while AI tools enhance clarity, flow, and efficiency, they raise concerns about over-reliance, originality, and ethical transparency. Advanced-year scholars valued citation improvements but were critical of originality-related functions. The study emphasizes responsible AI use, advocating for its role as a complementary, supportive tool to foster creativity and intellectual rigor. These findings contribute to the ongoing discourse on integrating AI responsibly into academic practices.

AI ethicsparaphraseplagiarism detectionacademic integrity
ArticleOpen access202527citation· Semantic Scholar

Application of machine learning assisted multi-variate UV spectrophotometric models augmented by kennard stone clustering algorithm for quantifying recently approved nasal spray combination of mometasone and olopatadine along with two genotoxic impurities: comprehensive sustainability assessment

BMC Chemistry · vol. 19 (1) · 98-98

The recent approval of the nasal spray combination of mometasone (MOM) and olopatadine (OLO) presents a significant analytical challenge, as only a single reported method exists for its determination, deviating from eco-friendly practices. This study addresses this critical gap by pioneering the application of machine learning techniques to develop robust UV spectrophotometric approach for the simultaneous quantification of MOM and OLO, along with two genotoxic impurities: 4-dimethylamino pyridine (DAP) and methyl para-toluene sulfonate (MTS). By simultaneously determining these highly concerning genotoxic impurities and active pharmaceutical ingredients, this method underscores its paramount significance in upholding rigorous pharmaceutical quality standards and safeguarding patient safety. Applying the multilevel-multifactor experimental design, the calibration set was meticulously chosen at five different concentrations, yielding 25 calibration mixtures with central levels of 4, 46.5, 2.5, and 3 µg/mL for MOM, OLA, MTS, and DAP, respectively. The key innovation lies in the strategic implementation of the Kennard-Stone Clustering Algorithm to create a robust validation set of thirteen mixtures, resolving the limitations of reported chemometric methods' random data splitting. This approach ensures unbiased evaluation across the full concentration space, improving the method's reliability and sustainability. The robustness of this approach was rigorously tested using five distinct chemometric models: principal component regression, classical least squares, partial least squares, genetic algorithm-partial least squares, and multivariate curve resolution-alternating least squares, demonstrating its broad applicability across diverse modeling techniques. All models successfully determined all components with excellent recovery, low bias-corrected prediction, and adequate limits of detection. The Greenness Index Spider Charts and the Green Solvents Selection Tool were used to choose environmentally conscious solvents. A comprehensive sustainability assessment employed six state-of-the-art tools, including the national environmental method index, complementary green analytical procedure index, analytical greenness metric, blue applicability grade index, carbon footprint analysis, and the red-green-blue 12 metrics. Favorable results across all metrics affirmed the method's eco-friendliness, real-world applicability, and cost-effectiveness, supporting sustainable development goals in pharmaceutical quality control processes.

mometasoneolopatadinenasal spray formulationgenotoxic impurities
ArticleOpen access202524citation· Semantic Scholar

Early attention-deficit/hyperactivity disorder (ADHD) with NeuroDCT-ICA and rhinofish optimization (RFO) algorithm based optimized ADHD-AttentionNet

Scientific Reports · vol. 15 (1) · 6967-6967

The ADHD detector analyzes behavioral, cognitive, or physiological data (e.g., EEG, eye-tracking, or surveys) to identify patterns associated with ADHD symptoms. This work offers a more sophisticated method of detecting ADHD by overcoming the main drawbacks of existing approaches in terms of data processing, detection accuracy, and computational time. The work is inspired by the fact that Deep Learning (DL) frameworks could transform the existing detection systems of ADHD. In the proposed framework, there is a new NeuroDCT-ICA module for the preprocessing of raw EEG data, which guarantees the elimination of noise and extraction of informative features. Moreover, the method introduces a novel RhinoFish Optimization (RFO) algorithm for selecting optimal features, which enhance the data processing capacity and the stability of the system. As a core of the approach, there is the ADHD-AttentionNet - the deep learning-based model aimed at improving the accuracy and confidence of ADHD identification. The model is validated with the standard metrics, and the performance of the model is outstanding as it has high accuracy of 98.52%, F-score of 98.26% and specificity of 98.16%. These outcomes show that the proposed model yields better accuracy in detecting ADHD related patterns.

attention deficit hyperactivity disorderADHD detectionEEG preprocessingdiscrete cosine transform