بحوث سكوبس — د. عبدالكاظم عبدالكريم الدريعي

Classification remains vital in cybersecurity since different data require different protection measures against threats. This paper compares the Decision Tree, K-Nearest Neighbor (KNN) and Logistic Regression algorithms in tackling cybersecurity datasets especially ones that are unbalanced. The performance of the models was evaluated with the help of Mean Absolute Error (MAE), Mean Squared Error (MSE), RMSE, R2 Score and Accuracy along with the Classification Report which gives precision, recall, F1-score and support of each class. From the table, it is evident that the Decision Tree classifier performs the best in terms of error rates hence obtaining the accuracy of 98. Part B: Multiple regression analysis of the comparative market analysis values, for clients the R2 Score attained was 09% and an R2 Score of 89. 56%. Despite fairly good results obtained by KNN, its accuracy was not equally effective across the minority classes. Relative to the other models, Logistic Regression, was the most accurate but it recorded the highest error rate. Based on that the results, it can be concluded that even though Decision Tree models turned out to be better for this purpose, KNN and the Logistic Regression could be better if they are to be optimized. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

The rapid growth in wireless communication demands has led to a surge in research on technologies capable of enhancing communication reliability, coverage, and energy efficiency. Among these, uncrewed aerial vehicles (UAV) and reconfigurable intelligent surfaces (RIS) have emerged as promising solutions. Prior research on using deep reinforcement learning (DRL) to integrate RIS with UAV concentrated on enhancing signal quality and coverage, but it ignored the challenges caused by electromagnetic interference (EMI). This article introduces a novel framework addressing the challenges posed by EMI from Gallium nitride (GaN) power amplifiers in RIS-assisted UAV communication systems. By integrating DRL with quadrature phase shift keying (QPSK) modulation, the proposed system dynamically optimizes UAV deployment and RIS configurations in real-time, mitigating EMI effects, improving signal-to-interference-plus-noise ratio (SINR), and enhancing energy efficiency. The framework demonstrates superior performance, with an SINR improvement of up to 6.5 dB in interference-prone environments, while achieving a 38% increase in energy efficiency compared to baseline models. Additionally, the system significantly reduces EMI impact, with a mitigation rate of over 70%, and extends coverage area by 35%. The integration of QPSK and DRL allows for real-time decision-making that balances communication quality and energy consumption. These results show the system’s potential to outperform traditional methods, particularly in dynamic and challenging environments such as urban, disaster recovery, and remote settings. © Copyright 2025 Ahmad et al. Distributed under Creative Commons CC-BY 4.0

In this paper, we introduce an innovative approach for extracting trajectories from a camera sensor in GPS-denied environments, leveraging visual odometry. The system takes video footage captured by a forward-facing camera mounted on a vehicle as input, with the output being a chain code representing the camera’s trajectory. The proposed methodology involves several key steps. Firstly, we employ phase correlation between consecutive video frames to extract essential information. Subsequently, we introduce a novel chain code method termed “dynamic chain code”, which is based on the x-shift values derived from the phase correlation. The third step involves determining directional changes (forward, left, right) by establishing thresholds and extracting the corresponding chain code. This extracted code is then stored in a buffer for further processing. Notably, our system outperforms traditional methods reliant on spatial features, exhibiting greater speed and robustness in noisy environments. Importantly, our approach operates without external camera calibration information. Moreover, by incorporating visual odometry, our system enhances its accuracy in estimating camera motion, providing a more comprehensive understanding of trajectory dynamics. Finally, the system culminates in visualizing the normalized camera motion trajectory. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

COVID-19 has affected millions of people around the world in the last three years. Despite widespread vaccination efforts, infections persist and definitive treatments remain elusive. Therefore, early and accurate detection of COVID-19 is critical to minimize invasive procedures and reduce mortality. Although radiographs and CT scans are commonly used for the diagnosis of COVID-19, electrocardiogram (ECG) images remain underutilized despite their widespread availability. This limited use can be attributed to the complex transformations required to process ECG data, which increase computational demands. This study proposes a novel hybrid deep learning model ECGNet-ViT for COVID-19 detection. The model combines the multi-scale feature extraction capabilities of GoogleNet (GNet) with Swish activation functions and densely connected layers, and then integrates it with Vision Transformer (ViT) to effectively capture long-range dependencies in classification tasks. This approach can efficiently analyze ECG data and accurately classify samples into five categories: normal, COVID-19, myocardial infarction (MI), previous myocardial infarction (PMI) and arrhythmia (AHB). Comprehensive experiments on a publicly available ECG datasets demonstrate the effectiveness of the proposed model, achieving 99.13% accuracy, 99.19% precision, 99.24% recall, and 99.22% F1 score. These results highlight the potential of the proposed model to provide reliable, non-invasive support in COVID-19 diagnosis based on ECG data. © 2025, Chulalongkorn University, Faculty of Fine and Applied Arts. All rights reserved.

This paper introduces a holistic and scalable framework for optimizing Large Language Models (LLMs) in distributed environments, addressing three critical challenges: computational efficiency, ethical fairness, and governance. As LLMs scale, issues, such as excessive resource consumption, fairness violations, and limited transparency, hinder their broader deployment in real-world applications. We propose a novel three-tier architecture that integrates topology-aware parallelism, communication-efficient gradient aggregation, and memory-aware rematerialization. Our implementation reduces training time by 38% and memory usage by 42% on a 512-GPU A100 cluster, without compromising accuracy. To promote fairness, we incorporate a real-time adversarial debiasing module that reduces demographic AUC gaps by over 60% across gender, ethnicity, and religion. For model interpretability, we introduce a symbolic explainability engine that converts attention weights into transparent rule-based explanations, achieving 89.2% user satisfaction and outperforming Grad-CAM and vanilla attention. Furthermore, a lightweight governance layer aligned with ISO/IEC 27001 and ISO/IEC 23894 standards ensures traceability, audit logging, and policy enforcement throughout the model lifecycle. We validate our framework across diverse datasets, including C4, WikiText-103, RealNews, and BookCorpus, demonstrating low-latency drift and consistent fairness across domains. Comparative benchmarks against DeepSpeed, FairScale, and Megatron-LM show superior throughput, energy efficiency, and transparency. This work advances the foundation for ethical, efficient, and regulation-compliant LLM deployment at scale. © The Author(s) 2025.

The supremacy of deep learning in artificial intelligence (AI) contexts, including image and speech recognition, computer vision, and medical imaging, among others, has established it as AI’s dominant approach. Several studies have been conducted on the use of deep learning in physiological signals, especially in ECG signals, in recent years, but there has been a lack of comprehensive review on the use of deep learning in ECG for biometric systems. This review is divided into two main sections: it provides a comprehensive bibliographic review of deep learning for ECG classification towards assisting in disease diagnosis in the first part while presenting an overview of the field, pioneers, and landmark studies. The second part offers comprehensive information on the subject, starting with the mathematical background of deep learning algorithms, the ECG signal processing, and the function of the heart. Using a PRISMA framework, 309 research papers were initially identified through specified keywords. After applying inclusion criteria, 90 articles were retained for detailed analysis, excluding 24 documents based on exclusion criteria EC1 and the remainder due to EC2. Key findings reveal that deep learning models achieve an average accuracy improvement of 10-15% over traditional methods, with convolutional neural networks (CNNs) and recurrent neural networks (RNNs) demonstrating superior performance in capturing complex ECG patterns. Through ECG databases, deep learning algorithms, assessment frameworks, metrics, and code availability, this review designs a systematic view from different perspectives to highlight the trends, challenges, and opportunities of deep learning for ECG arrhythmia classification. This paper’s goal is to contribute to the knowledge of both new and experienced researchers and practitioners in the field so that they can learn and understand the various processes involved in ECG signal processing using deep learning. © 2024, Chulalongkorn University, Faculty of Fine and Applied Arts. All rights reserved.

Cardiovascular diseases, which are currently the major causes of death globally, can be largely ameliorated through early detection and categorization. Electrocardiogram (ECG) tests have emerged as widely employed, low-cost and non-invasive procedures for evaluating electrical activities of the heart and diagnosing cardiovascular ailments. In this research, by using deep learning techniques to detect specific cardiac disorders like cardiac myocardial infarction(MI), arrhythmia, past history of myocardial infarction(PMI) and normal ECG patterns on a dataset containing patients with heart disease. We propose ECGConVT framework that combines Convolutional Neural Network (CNN) module for extracting local features, and Vision Transformer (ViT) module for capturing global features. The final classification is achieved by combining the two using Multilayer Perceptron (MLP) module. The experimental results indicate promise of ECGConVT in ECG image classification where it outperforms other approaches showing an average accuracy of 98.5%, F1-score: 98.7%, Recall: 98.8% and Precision: 98.5%. In order to meet the practical needs of clinical applications, we implemented a lightweight post-processing step to reduce the size of the model. © 2013 IEEE.