Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Ahmed A. Elngar has a total of 289 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Neural Depths

    by B Mohanraj, M Priya, Ahmed A. Elngar, P. Sachidhanandam, S Lalithambikai

    The world oceans cover over 70% of the Earth's surface and form one of the least explored and very intricate destinations. Such strategic value of the waters, ranging everywhere between international trade lanes and naval warfare and defense [ 1 ] to the ecological balance and discovery of resources, has given underwater surveillance the status of a mandatory field of modern research. The underwater department is, however, different since it has its own issues unlike a ground-dwelling environment or an aerial environment. It is insufficient because of the challenges of traditional surveillance means due to poor visibility, complex marine acoustics, unpredictable ocean currents, and vibrant biological presence.

  • Digital Twins and Intelligent Manufacturing for Breast Cancer Devices

    by Farah Hesham, Mohammed Mostafa, Mohammed Abdalla, Ahmed A. Elngar

    This chapter surveys how digital twins (DTs) and intelligent manufacturing are reshaping the lifecycle of breast-cancer medical devices through Internet of Things (IoT)-driven data flows. We review architectures that link design, production, clinical use, and post-market surveillance via a “digital thread,” enabling continuous feedback between factory and clinic. On the manufacturing floor, we analyze model-based design, in-line sensing, and edge analytics for process control, traceability, and rapid changeover, covering modalities such as diagnostic wearables, imaging adjuncts, biopsy tools, and implantable systems. For clinical deployment, we examine patient- and fleet-level DTs that integrate device telemetry with electronic health record (EHR)-adjacent data to support calibration, predictive maintenance, and personalized therapy support. The chapter proposes a reference framework spanning edge–fog–cloud layers, data models, and control loops, and synthesizes evidence on benefits (reduced defect rates, shorter release cycles, and improved reliability) alongside open challenges in interoperability, verification/validation of twin models, cybersecurity, privacy, and regulatory alignment. We also outline key performance indicators (KPIs) for assessing maturity, process capability, time-to-release, real-world performance drift, and post-market vigilance responsiveness. Future directions include physics-informed and hybrid (data + mechanistic) twins, standardized digital threads across suppliers, federated analytics for privacy-preserving learning, and zero-trust security patterns. Taken together, the chapter offers a practical roadmap for translating IoT-enabled DTs from promising pilots to scalable, compliant manufacturing and clinical value in breast-cancer device ecosystems.

  • Interpretable Deep Transfer Learning Models for Breast Cancer Classification

    by Sara Hamed, Rasha A. Ali, Asmaa Magdy Salama, Ahmed A. Elngar, El-Sayed M. El-Horbaty

    Breast cancer is still the most prevalent type of cancer leading to death for women around the globe; thus, the requirement for an effective and efficient early detection and diagnostic system remains necessary. Different types of deep learning models, such as convolutional neural networks (CNNs) and vision transformer (ViT) architectures, have successfully been used to classify breast cancer automatically using various imaging modalities, such as ultrasound, mammography, and histopathology. However, the “black-box” nature of these models creates a barrier to implementation in clinical practice because clinicians will only trust a decision support system if it is transparent to them, allowing them to be assured of patient safety. This chapter describes some of the latest advancements (2020–2026) in interpretable deep transfer learning (DTL) techniques to classify breast cancer. This includes a comparison of the most commonly used models (EfficientNet, ResNet, DenseNet) and hybrid CNN-ViT models, as well as the performance and the ways that transfer learning has been used in these models. The explainable artificial intelligence (XAI) methods of Grad-CAM, LIME, and Sensitivity maps are discussed as methods to achieve visual and feature-level interpretability. We have found documentation for multiple studies reporting classification accuracies above 99% for hybrid models using attention mechanisms and other improvements on benchmark datasets. An integrated model framework that contains deep extracted features, transfer learning, XAI modules, and clinical decision support will be proposed as well. This chapter not only describes the accompanying integrated model frameworks but will also consider ethical issues and regulatory standards (FDA & MDR), the challenges of applying research-study findings to the real world, and future research directions. This chapter will describe some of the practical issues associated with integrating interpretable deep learning models into an ontology in a transparent way while ensuring they adhere to all applicable regulations.

  • AI-Driven Epidemic Forecasting

    by Shrabanti Kundu, Utpal Ghosh, Ahmed A. Elngar

    This chapter presents an artificial intelligence (AI)-driven predictive framework designed to strengthen epidemic forecasting and support public health intelligence. By integrating machine learning, deep learning, and Internet of Things (IoT)-enabled data acquisition, the system synthesizes heterogeneous data sources, including clinical health records, environmental parameters, social media activity, and real-time sensor feeds, to deliver timely and accurate outbreak predictions. The chapter provides a comprehensive review of existing epidemic forecasting approaches, spanning traditional compartmental models (SIR and SEIR), statistical models such as ARIMA and Exponential Smoothing and classical machine learning (ML) techniques. While these models have provided some important insights, static assumption, and restrictive adaptability, and lack of real-time responsiveness limit their predictive function, especially in the case of newer pathogens. To overcome such limitations, in the AI approach presented herein deep neural networks, transformer-based, and hybrid ensemble learning algorithms are employed for better forecasting accuracy and adaptability. The methodology follows a structured pipeline encompassing multi-source data collection, feature optimization using both statistical and AI-based methods, and model training on real-world epidemiological datasets. Metrics such as R², Mean Absolute Error (MAE), and Root Mean Squared Error (RMSE) are used to evaluate the model performance, thereby revealing that the proposed methods have led to improvements in the predictive accuracy, sensitivity, and computational efficiency as opposed to conventional methods.

  • Different Genomic Representations of Novel Pathogens Based on Signal-Processing Algorithms with a Case Study on COVID-19

    by Rabeb Touati, Mohamed Touati, Faouzi Benzarti, Vishal Kumar, Maher Kharrat, Ahmed A. Elngar

    Coronaviruses are a type of frequent RNA virus responsible for digestive and respiratory infections in animal and human genomes. A novel coronavirus, later named COVID-19, appeared and spread worldwide, which led to the World Health Organization (WHO) to declare it a pandemic in March 2020. The SARS-CoV-2 genome, the virus responsible for the COVID-19, has a size of 29,903 nucleotides, and its genetic makeup is composed of 11 functional open reading frames (ORFs). This chapter proposes comparison algorithms to find the origin of SARS-CoV-2 using digital genomic signatures and signal processing algorithms. As a result, the five closest genomes related to SARS-CoV-2 are RaTG13, Pangolin-CoV-PCoV_GX-P2V, Pangolin CoV MP789, bat-SL-CoVZC45, and bat-SL-CoVZXC210. The spectra correlation results are the existing of groups and each group presents more similar viruses: (1) We have A, B, C, D, V, and X viruses; (2) we have I, J, K, L, and U viruses; (3) S, Y, Z, R, Q, and H viruses; and (4) G and O viruses. These phylogenetic tree results, using a classification based on numerical DNA sequences, reflect the performance of our algorithms. Our methods can be used to solve biological problems.

  • Handbook of Smart Healthcare

    by Ahmed A. Elngar, Sonal Trivedi, S. David Samuel Azariya, Diego Oliva, Valentina E. Balas

    In an era where healthcare innovation and technological advancement converge at unprecedented speed, the integration of intelligent manufacturing into medical

  • Open Access

    An optimized approach for brain tumor classification using MobileNet and enhanced Newton-Raphson optimizer

    by Essam H. Houssein, Bahaa El-Din Helmy, Ahmed A. Elngar, Ahmed Taha, Hassan Shaban

    This article introduces an enhanced version of the Newton-Raphson-based optimizer (I-NRBO) with a complex optimization technique termed the Local Escaping Operator (LEO). This method effectively balances solution variety with convergence speed, enhancing optimization effectiveness and reducing the likelihood of local optima. The research presents an advanced and efficient model that utilizes a pre-trained deep learning architecture, MobileNet, with the I-NRBO algorithm to enhance the classification accuracy of brain tumors. The I-NRBO method is evaluated over CEC-2022, the results showed that I-NRBO outperformed the traditional NRBO and other leading algorithms in terms of statistical convergence and various criteria. In the subsequent study, the proposed I-NRBO is applied to optimize hyperparameters in the MobileNet model, specifically for brain tumor classification. The I-NRBO-MobileNet model is assessed against many previous research and alternative deep learning architectures with comparable parameters, including VGG16, ResNet50, InceptionV3, and MobileNet as baseline models. The experimental findings reveal that the I-NRBO-MobileNet model outstanding performance, with an accuracy of 99.24%, a precision of 99.25%, a recall of 99.24%, an AUC of 99.95%, and a specificity of 99.70%. Results demonstrate that the suggested I-NRBO-MobileNet model can accurately classify brain cancers.

  • Open Access

    A comprehensive review of artificial intelligence and Internet of Things integration based plant disease detection for sustainable agriculture

    by Ahmed A. Elngar, Howaida Allam, Hamdi A. Mahmoud

    Detecting crop diseases early and responding promptly is vital for protecting agricultural productivity. It also helps maintain the quality and quantity of yields and reduces the risk of disease transmission to humans and livestock. Effective disease management is therefore critical to ensuring both global and local food security. However, traditional methods often based on visual inspection and delayed human judgment, are typically insufficient for identifying diseases at an early stage. Recent developments in Artificial Intelligence (AI) and the Internet of Things (IoT) offer new opportunities to address these challenges. By integrating IoT sensor networks with AI techniques such as machine learning and deep learning, it becomes possible to monitor plant health in real time and detect diseases with greater accuracy. This review explores the strengths and limitations of current AI-enabled IoT solutions in agriculture. It highlights how these systems leverage large-scale data and advanced image processing to outperform conventional methods in terms of speed, precision, and efficiency. Such improvements can significantly reduce crop losses and support more sustainable agricultural practices. Finally, the paper reviews key research trends, identifies current challenges, and outlines future directions in the field. It emphasizes the transformative potential of smart agriculture in advancing plant disease management and promoting environmentally responsible food production. This systematic review was conducted in strict accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines, synthesizing a final selection of 152 peer-reviewed papers. The overarching aim is to critically evaluate and map literature published between 2015 and 2026 using a systematic approach that addresses the integration of the IoT, AI, Machine Learning (ML), Deep Learning (DL), Convolutional Neural Networks (CNN), sensor technologies, and sustainable agricultural practices in the context of plant disease detection.

  • Distinguishing Security Monitoring From Surveillance Within Legal Boundaries

    by Mahmoud Abu Saima, Ahmad A. Elngar

    Within the boundaries of law, a clear distinction is made between security monitoring and surveillance. It does a good job of explaining the conceptual, functional, and operational differences between protective monitoring of systems, assets, and people on the one hand, and surveillance practices, which are about systematic observation of persons or behavior, on the other. The chapter breaks down the way in which purpose, extent, proportionality, and data handling determine whether these practices fall under privacy and data protection laws. It also talks about consent, transparency, necessity, and minimization as the main principles that security monitoring needs to follow to be lawful. The chapter through law and case studies draws attention to the danger of function creep, over-collection of data, and misuse of monitoring tools. It offers a roadmap for organizations to develop their security measures both to be effective and at the same time comply with rights, laws, and ethics.

  • Regulatory Readiness of Artificial Intelligence-Based Security Systems

    by Ahmed A. Elngar, Mahmoud Mahfuri

    This chapter looks into how regulatory ready are artificial intelligence (AI) based security systems that have been deployed in regulated and high-risk environments. It considers the extent to which preparedness is influenced by governance structures, documentation, risk assessments, and operational controls throughout the system lifecycle. The chapter delves into the alignment with data protection, cybersecurity, and sector-specific regulations, and highlights the requirements for accountability, transparency, human oversight, and auditability. The chapter particularly focuses on issues raised by changing regulations, jurisdictional requirements, and reliance on third- party vendors. By combining legal expectations with security practices, the chapter offers a conceptual framework for evaluating regulatory readiness, identifying gaps, and thus helping organizations to foresee compliance requirements, lower the risk of regulatory actions, and implement AI-enabled security systems in a responsible manner.