Egyptian Informatics Journal最新文献

{"title":"ADAMAEX—Alzheimer’s disease classification via attention-enhanced autoencoders and XAI","authors":"Doorgeshwaree Bootun ,&nbsp;Muhammad Muzzammil Auzine ,&nbsp;Noor Ayesha ,&nbsp;Salma Idris ,&nbsp;Tanzila Saba ,&nbsp;Maleika Heenaye-Mamode Khan","doi":"10.1016/j.eij.2025.100688","DOIUrl":"10.1016/j.eij.2025.100688","url":null,"abstract":"<div><div>To bring a new contribution in the area of classification of Alzheimer’s Disease (AD) we introduce a deep learning model, ADAMAEX, which is based on a convolutional autoencoder with four convolutions in the encoder part and a Squeeze and Excitation block for channel attention applied after each convolution. Additionally, we utilised fully connected layers (dense layers) for AD image classification. To conduct our study, we specifically chose axial brain scans acquired through sMRI in T2-weighted mode from the ADNI database, which were augmented using colour jitter, rotations, and flipping techniques. Before feeding the images to the model, we applied pre-processing steps such as re-sampling, normalisation, Contrast-Limited Adaptive Histogram Equalisation (CLAHE), and sharpening using the Unsharp Mask technique. For visualisation, we integrated Grad-CAM, an Explainable AI (XAI) technique, to highlight the brain regions responsible for the model’s classification decisions, a method underutilised by other authors in the context of AD classification. This model achieved an impressive accuracy of 96.2% and shows great promise for adoption in the medical sector, providing valuable assistance to doctors in validating their predictions based on brain scans.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"30 ","pages":"Article 100688"},"PeriodicalIF":5.0,"publicationDate":"2025-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144084631","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Cloud based collaborative data compression technology for power Internet of Things","authors":"Qiong Wang ,&nbsp;Yongbo Zhou ,&nbsp;Jianyong Gao","doi":"10.1016/j.eij.2025.100696","DOIUrl":"10.1016/j.eij.2025.100696","url":null,"abstract":"<div><div>To address the challenge of explosive data growth in power IoT systems, this study develops a cloud-edge collaborative multi-task computing framework for efficient compression of heterogeneous data. The proposed system builds upon a “microservice-containerization-Kubernetes” architecture that enables parallel processing of multi-source IoT data collected through perception layer devices. At the edge layer, a hybrid performance ontology algorithm first integrates diverse data sources, followed by a two-stage compression approach: wavelet transforms perform initial data aggregation, while tensor Tucker decomposition enables secondary compression for optimized data reduction. Experimental results demonstrate the framework’s effectiveness in maintaining stable IoT network operations while achieving compression ratios below 40%, significantly improving upon traditional methods in both efficiency and reliability for power IoT applications.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"30 ","pages":"Article 100696"},"PeriodicalIF":5.0,"publicationDate":"2025-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144070478","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A strong algorithm for randomly hiding a secret files inside true color image using large primary secret key","authors":"Aamer Tahseen Suhail,&nbsp;Harith Ghanim Ayoub,&nbsp;Ammar Abdulmajeed Gharbe","doi":"10.1016/j.eij.2025.100692","DOIUrl":"10.1016/j.eij.2025.100692","url":null,"abstract":"<div><div>Hiding in image files is one of the widespread techniques to achieve information security, as it has proven its efficiency.</div><div>Because of the hiding technology has become the best way to maintain the confidentiality of private information, the attackers have doubts about any translated media, such as various image files, because the possibility of containing important confidential data, then trying to use different methods and algorithms to extract its containing. Therefore, the traditional methods of regular and sequential hiding have become at a steady pace within the media, weak and easy to detect by the attackers, so the trend was to find a new algorithm that adopts the method of random hiding scattered within those media.</div><div>During the research, a method was developed to hide secret files inside true color images, by distributing the content of it within the image elements in a way that makes the distance between the locations of the image elements in irregular and not fixed space, based on an algorithm random distribution depend on a large prim secret key that agreed upon paging coefficient among the authorized persons, which makes it difficult to extract the hiding locations of those parts within them using image analysis methods, or through statistical analyzes that used in this field.</div><div>This method succeeded in hiding the secret files without causing a state of distortion to the original image, or the possibility of noticing the changes that occurred in it as a result of the hiding process, and these files were easily retrieved without losing any of their components, in addition to the fact that this extracting process takes place without require the original cover image or need to create a table of shows hiding locations.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"30 ","pages":"Article 100692"},"PeriodicalIF":5.0,"publicationDate":"2025-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143936736","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Design and optimization of human-machine interaction interface for the intelligent Internet of Things based on deep learning and spatial computing","authors":"Wencong Wang,&nbsp;Ke Wang,&nbsp;Hui Du","doi":"10.1016/j.eij.2025.100685","DOIUrl":"10.1016/j.eij.2025.100685","url":null,"abstract":"<div><div>The Intelligent Internet of Things (IoT) is transforming interactions with smart devices, especially in a home environment, with lighting, security, and entertainment systems. Designing user-friendly interfaces for IoT systems presents difficulties, particularly for individuals with severe disabilities such as Amyotrophic Lateral Sclerosis (ALS), spinal cord injuries, and cerebral palsy. Existing user interfaces frequently restrict the capacity of individuals, particularly those with severe disabilities, to operate smart home gadgets efficiently. The study proposes that the NeuroSpatialIOT system solve this problem by combining 2D spatial mapping, deep learning, and eye tracking. The system’s innovative approach accurately interprets user intent through natural eye gaze and provides relevant controls based on the user’s viewpoint and environment. NeuroSpatialIOT leverages deep learning to gather data on eye movements, the 2D spatial configuration of the room, and user objectives. NeuroSpatialIOT functions by tracking eye movements, comprehending the room’s 2D layout, and employing deep learning to predict user intentions. The system then displays appropriate controls in the user’s field of view, enabling intuitive interaction with IoT devices. Testing on non-disabled and severely disabled people yielded positive findings. Non-disabled participants scored 88.9% and disabled individuals 91.5%, indicating great system usability. Subsequently took 40% less time for non-impaired users to complete tasks and 60% less for disabled users. With NeuroSpatialIOT, altering room temperature or illumination takes 10–15 s instead of 30–45. The results show that the system can improve autonomy and quality of life for varied users in IoT-enabled settings by making home operations easier to handle.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"30 ","pages":"Article 100685"},"PeriodicalIF":5.0,"publicationDate":"2025-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143928599","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhancing IoT security and healthcare data protection in the metaverse: A Dynamic Adaptive Security Mechanism","authors":"Saima Siraj Qureshi ,&nbsp;Jingsha He ,&nbsp;Nafei Zhu ,&nbsp;Ahsan Nazir ,&nbsp;Juan Fang ,&nbsp;Xiangjun Ma ,&nbsp;Ahsan Wajahat ,&nbsp;Faheem Ullah ,&nbsp;Sirajuddin Qureshi ,&nbsp;Sahroui Dhelim ,&nbsp;Muhammad Salman Pathan","doi":"10.1016/j.eij.2025.100670","DOIUrl":"10.1016/j.eij.2025.100670","url":null,"abstract":"<div><div>In the rapidly evolving landscape of the Metaverse, the synergistic integration of the Internet of Things (IoT) and Digital Twins (DT) represents a revolutionary paradigm shift, seamlessly bridging the real and virtual worlds. While this innovative convergence offers unprecedented potential, it also exposes a broader spectrum of security vulnerabilities that challenge conventional approaches. This research aims to fortify the multifaceted ecosystem of the Metaverse, with a particular emphasis on securing IoT healthcare data. Ensuring the protection of health information within the extensive network of interconnected devices in the Metaverse is paramount. Addressing this critical need, we introduce the Dynamic Adaptive Security Mechanism (DASM), an advanced Artificial Intelligence (AI)-driven framework meticulously crafted to enhance security adaptively. DASM operates as a comprehensive and real-time defensive layer, continuously recalibrating its strategies to reinforce the security matrix for both IoT and Digital Twins. This study provides a detailed examination of the foundational architecture of DASM and its AI-driven adaptive processes. We elucidate its pivotal role in strengthening the security framework within the complex terrain of the Metaverse. Through rigorous testing and validation using the IoT healthcare security dataset, the Random Forest model emerges as the top performer, achieving near-perfect metrics, including a Matthews Correlation Coefficient (MCC) of 0.9989 and superior Balanced Accuracy, while also offering reduced training and inference times compared to the LSTM model. Although the LSTM model demonstrates strong accuracy, the ensemble approach of Random Forest balances computational efficiency and performance. The DASM framework sets a new benchmark in IoT security, offering a scalable and effective solution with significant implications for the future of Metaverse applications.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"30 ","pages":"Article 100670"},"PeriodicalIF":5.0,"publicationDate":"2025-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143924300","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Ransomware detection and family classification using fine-tuned BERT and RoBERTa models","authors":"Amjad Hussain ,&nbsp;Ayesha Saadia ,&nbsp;Faeiz M. Alserhani","doi":"10.1016/j.eij.2025.100645","DOIUrl":"10.1016/j.eij.2025.100645","url":null,"abstract":"<div><div>Integrating Internet of Things (IoT) technologies in healthcare has revolutionized patient care, enabling real-time monitoring, predictive analytics, and personalized treatments. However, it presents significant challenges that must be addressed to ensure secure and reliable implementation. IoT devices in healthcare, such as remote patient monitors, are often constrained by limited computational power, making them vulnerable to sophisticated cyberattacks, including ransomware. In 2017 the WannaCry ransomware attack disrupted many National Health Service facilities in the United Kingdom and emphasized the critical need for robust cybersecurity measures. The lack of standardization across IoT devices creates interoperability issues and complicates data transfer between medical devices and healthcare systems. This research explores these challenges and proposes a novel approach using hyperparameter-optimized transfer learning-based models, Bidirectional Encoder Representations from Transformers (BERT), and a Robustly Optimized BERT Approach (RoBERTa), to not only detect but also classify ransomware targeting IoT devices by analyzing dynamically executed API call sequences in a sandbox environment. A total of 3300 samples from 10 ransomware families including 300 benign cases are analyzed dynamically in a sandbox environment. The newly created dataset is then preprocessed and fed to the BERT and RoBERTa models for training. The BERT achieved 95.60% accuracy with a minimal loss of 0.1650 while the RoBERTa achieved 94.39% accuracy with 0.1948 loss in classifying ransomware families. These results indicate that the proposed approach is game-changing in the classification of previously unidentified behavioral patterns inside ransomware and enhances the ability to tackle newly developing threats. By leveraging the dynamic analysis with API call sequences in a correct format, and training hyperparameter-optimized transformer learning-based models, the methodology efficiently captures behavioral patterns unique to ransomware. The research provides a scalable framework for integrating advanced detection mechanisms into real-world healthcare IoT systems, enhancing their resilience against cyber threats.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"30 ","pages":"Article 100645"},"PeriodicalIF":5.0,"publicationDate":"2025-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143924299","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"LPBSA: Pre-clinical data analysis using advanced machine learning models for disease prediction","authors":"Dana R. Hamad ,&nbsp;Tarik A. Rashid","doi":"10.1016/j.eij.2025.100690","DOIUrl":"10.1016/j.eij.2025.100690","url":null,"abstract":"<div><div>Diabetes, COVID-19, and heart disease pose significant global health challenges. The current study introduces an optimization algorithm, Learner Performance-Based Behavior with Simulated Annealing (LPBSA), integrated with Multilayer Perceptron (MLP) as a neural network technique to improve disease prediction accuracy. The algorithm was tested on six preclinical datasets (one is related to diabetes, two are related to heart disease, and three are related to COVID-19). In addition to LPBSA-MLP, other optimization algorithms, including Fitness Dependent Optimizer (FDO), the original Learner Performance-Based Behavior (LPB), were independently combined with MLP. Furthermore, all three algorithms were integrated with a Cascading Multilayer Perceptron (LPBSA-CMLP, FDO-CMLP, LPB-CMLP) to enhance the convergence speed and learning capability. This allowed for a comprehensive comparison across diverse algorithmic configurations and enabled the identification of the most efficient model for disease prediction. The proposed LPBSA-MLP model achieved 100% accuracy on four data sets and at least 99.31% on the others. Further metrics confirm its performance: sensitivity and specificity values reached 100%, and Mean Square Error (MSE) values ranged from 0.0008 to 0.003. When benchmarked against models trained with FDO-MLP, LPB-MLP, and other standard optimizers, LPBSA-MLP consistently outperformed them in terms of both classification performance and convergence speed. These findings indicate the effectiveness of LPBSA in enhancing predictive modeling for critical health conditions.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"30 ","pages":"Article 100690"},"PeriodicalIF":5.0,"publicationDate":"2025-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143907507","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Towards lightweight model using non-local-based graph convolution neural network for SQL injection detection","authors":"Duc-Chinh Nguyen ,&nbsp;Manh-Hung Ha ,&nbsp;Manh-Tuan Do ,&nbsp;Oscal Tzyh-Chiang Chen","doi":"10.1016/j.eij.2025.100684","DOIUrl":"10.1016/j.eij.2025.100684","url":null,"abstract":"<div><div>SQL injection poses serious threats to web applications and databases by enabling unauthorized access and data leakage. To address this issue, we propose a unique graph network, an innovative topology not explored previously for SQL injection detection. SQL statements are nodes, and their connections form edges in the graph. We introduce three graph CNN models, including a graph classification model with a two-layer Graph Convolutional Network (GCN), a graph classification model leveraging a non-local graph convolution derived from a 1x1 convolution, supplanting the original 1x1 convolution, and a modified non-local-block module by substituting the 1x1 convolution layers in the non-local architecture with GCN. The proposed models exhibit accuracy above 99% and inference times under 1 ms on two datasets. In comparison with traditional 22 models, our models using GCN demonstrate superior computation efficiency, parameter reduction, accuracy enhancement, and the advantage of handling input sequences of any length, underlining their potential in real-world cybersecurity systems, especially in effective SQL injection detection and mitigation strategies.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"30 ","pages":"Article 100684"},"PeriodicalIF":5.0,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143892375","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Intrusion detection algorithm of wireless network based on network traffic anomaly analysis","authors":"Xiangqian Nie,&nbsp;Jiao Xing,&nbsp;Qimeng Li,&nbsp;Fan Xiao","doi":"10.1016/j.eij.2025.100689","DOIUrl":"10.1016/j.eij.2025.100689","url":null,"abstract":"<div><div>Due to the openness and sharing nature of wireless networks, they are vulnerable to various network attacks. To promptly identify and mitigate abnormal behaviors while ensuring normal operation and security, this paper proposes an algorithm for detecting compromised nodes in wireless networks based on network traffic anomaly analysis. In the proposed detection architecture, a network traffic data acquisition module mines and reconstructs real-time traffic data from wireless nodes, removing redundant information. The processed data is then fed into an anomaly analysis module, where abnormal traffic features are extracted and dimensionality-reduced via a stacked autoencoder to form standardized anomaly profiles. These features are analyzed by an intrusion detection module combining particle swarm optimization and support vector machine algorithms. Experimental results demonstrate that the algorithm efficiently extracts traffic anomalies, accurately detects attack duration and traffic volume changes in compromised nodes, and maintains a false detection rate below 6 %.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"30 ","pages":"Article 100689"},"PeriodicalIF":5.0,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143881718","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Design and hardware implementation of LED block cipher for vehicles keyless entry systems","authors":"Ayoub Mhaouch ,&nbsp;Wafa Gtifa ,&nbsp;Abdesslem Abdeali ,&nbsp;Anis Sakly ,&nbsp;Mohsen Machhout","doi":"10.1016/j.eij.2025.100687","DOIUrl":"10.1016/j.eij.2025.100687","url":null,"abstract":"<div><div>Security is paramount in vehicle keyless entry systems, as they are increasingly targeted by various attacks, including relay attacks, theft, and espionage. Keyless entry systems are particularly vulnerable to relay attacks, where attackers intercept and amplify the signal from the key fob, granting unauthorized access to the vehicle. This compromises the integrity of the system, emphasizing the need for robust encryption mechanisms to prevent unauthorized access and safeguard sensitive vehicle data. In this work, we propose an optimized hardware design for the Light Encryption Device (LED) cipher, aimed at enhancing both the security and efficiency of keyless entry systems. The proposed security system is evaluated using security metrics such as NPCR, UACI, entropy, and correlation analysis, demonstrating its robust protection against potential attacks. The obtained results show that the proposed hardware implementation delivers higher efficiency and enhanced security compared to existing designs, making it a promising solution for securing keyless vehicle entry systems. The real-world test scenarios assess the performance of the proposed hardware system, demonstrating its effectiveness in terms of execution time, power consumption, and battery drain time across different platforms, including Dual Core ARM Cortex-A9 and Zynq XC7Z020. The results reveal that the proposed designs offer improved efficiency and security, positioning them as a viable solution for safeguarding keyless entry systems against unauthorized access. This work underscores the potential of lightweight cryptography to tackle both security and performance challenges in modern automotive systems, ensuring the safety and integrity of vehicle access control.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"30 ","pages":"Article 100687"},"PeriodicalIF":5.0,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143886873","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}