Safwan Mahmood Al-Selwi is a Senior Lecturer in the Faculty of Business & Technology at the University of Cyberjaya (Malaysia). He previously served as a Research and Teaching Assistant at the Department of Computing, Universiti Teknologi PETRONAS (UTP). He holds a PhD in Information Technology from UTP (Malaysia), an MSc in Computer Applications from Bangalore University (India), and a BEng in Software Engineering from Taiz University (Yemen). He has over ten years of experience in academic institutions and industry. His research interests include artificial intelligence, deep learning, computer vision, metaheuristic algorithms, and optimization, and his industry experience includes Android and web application development.
EDUCATION
PhD in IT - Universiti Teknologi PETRONAS (Malaysia).
MSc in Computer Applications - Bangalore University (India).
BEng in Software Engineering - Taiz University (Yemen).
RESEARCH, TEACHING, or OTHER INTERESTS
Artificial Intelligence, Computer Science
23
Scopus Publications
1777
Scholar Citations
12
Scholar h-index
12
Scholar i10-index
Scopus Publications
Explainable deep learning-based lung cancer diagnosis using clinically-guided local interpretable model-agnostic explanations Shahab Ul Hassan, Said Jadid Abdulkadir, Hitham Seddig Alhussian, Abdul Muiz Fayyaz, Safwan Mahmood Al-Selwi, Unsa Khan, Ahmed Omer Ahmed Ismail Scientific Reports, 2026 Lung cancer remains one of the leading causes of cancer-related deaths worldwide, highlighting the urgent need for accurate and interpretable diagnostic tools. While deep learning (DL) models have achieved strong results in medical image classification, their opaque decision-making process remains a barrier to clinical adoption. This study proposes an adaptive superpixel perturbation-based local interpretable model-agnostic explanations (ASP-LIME), a novel explanation framework designed to generate faithful and localized interpretations of DL predictions, providing insights into the model's decision-making process. The proposed approach improves upon the original local interpretable model-agnostic explanations method by introducing adaptive superpixel segmentation, stratified perturbation strategies, lung region masking, and post-processing enhancements tailored for medical imaging. The proposed framework is applied to a lung cancer classification task using a custom-designed convolutional neural network, MedDeepNet, as the predictive model. Experimental results on a publicly available lung image dataset demonstrate that MedDeepNet achieves 99.84% accuracy, 99.66% recall, 99.82% precision, 99.74% specificity, and a 99.74% F1-score. ASP-LIME produces high-fidelity explanations with strong localization to pathological regions, achieving scores of 0.0300 for deletion, 0.9622 for insertion, and 0.9661 for Area Between Perturbation Curves (ABPC), surpassing typical benchmarks for interpretability methods. The findings demonstrate that the proposed framework offers consistent and interpretable explanations that enhance understanding of model decisions in medical imaging applications.
The role of advanced machine learning in COVID-19 medical imaging: A technical review Abdul Muiz Fayyaz, Said Jadid Abdulkadir, Shahab Ul Hassan, Safwan Mahmood Al-Selwi, Ebrahim Hamid Sumiea, Lareib Fatima Talib Results in Engineering, 2025 Medical images such as CXR, ultrasound, and CT scans are crucial for the rapid and accurate diagnosis of COVID-19. However, the diagnostic process is often challenging, time-consuming, and prone to human error. Although machine learning (ML) techniques have shown great promise in addressing these challenges, there remains a need for a comprehensive overview of how these methods contribute to automated COVID-19 detection. This article provides a technical overview of ML techniques used for identifying and categorizing COVID-19 in medical images. It focuses on approaches such as Deep Learning (DL) algorithms and Transfer Learning, which have demonstrated significant potential in developing automated, accurate COVID-19 detection systems. The study reviews key research on ML-based COVID-19 classification, focusing on stages like preprocessing, segmentation, feature extraction, transfer learning, feature selection, and classification. It evaluates various methodologies, emphasizing advancements in efficiency and accuracy. The review also compares different ML and DL techniques, offering insights into their contributions to improving COVID-19 detection, diagnosis, and prediction. This synthesis aims to provide a comprehensive understanding of current AI-based approaches, supporting researchers, healthcare professionals, governments, and policymakers in enhancing the use of ML for COVID-19 classification. • Comprehensive review of machine learning applications for COVID-19 detection, encompassing diverse methodologies and tools. • Comparative analysis of machine learning and deep learning models for COVID-19 classification, diagnosis, and prediction. • Synthesis of significant research contributions, providing insights into advancements in COVID-19 image classification. • Detailed discussion on automated COVID-19 detection using CT, ultrasound, and X-ray imaging techniques. • Identification of future research opportunities and discussion of public datasets commonly used in COVID-19 classification studies. • Technical analysis of state-of-the-art techniques: - Preprocessing: Analysis of advanced noise reduction techniques for enhancing image quality. - Segmentation: Review of algorithms for isolating lung areas in medical images. - Feature Extraction: Examination of methods for identifying unique COVID-19 characteristics. - Feature Selection: Overview of techniques for selecting the most relevant features. - Classification: Review of models for accurate COVID-19 detection and severity categorization.
Deep deterministic policy gradient - model-agnostic meta-learning framework: Efficient adaptation in continuous control tasks Ebrahim Hamid Sumiea, Said Jadid Abdulkadir, Hitham Seddig Alhussian, Safwan Mahmood Al-Selwi, Alawi Alqushaibi, Mohammed Gamal Ragab, Abdul Muiz Fayyaz Results in Engineering, 2025 Deep reinforcement learning (DRL) demonstrates superior performance in continuous control tasks. However, extensive training across a variety of environments is frequently necessitates extensive training. This manuscript presents Meta-DDPG-MAML, which combines the Deep Deterministic Policy Gradient (DDPG) methodology with Model-Agnostic Meta-Learning (MAML) to augment both adaptability and efficiency. By incorporating meta-learning principles into the DDPG actor-critic architecture, the proposed approach facilitates swift adaptation utilizing minimal data. Empirical results from benchmark environments -LunarLanderContinuous-v2, BipedalWalker-v3, and Pendulum-v1- reveal that Meta-DDPG-MAML consistently surpasses DDPG in several performance metrics. In the LunarLander, it attains maximum returns approaching 200, effectively doubling the performance of DDPG while concurrently stabilizing episode durations. In the BipedalWalker, it exceeds 250 returns by the 5000th episode, whereas the DDPG plateaus below 300. In the Pendulum, both methodologies stabilize around -200. However, the Meta-DDPG-MAML framework exhibits a more consistent critic loss, averaging -52.5 compared to the fluctuations observed with DDPG. Across all tasks, Meta-DDPG-MAML achieves superior episode return stability, maintains higher critic loss consistency, and improves policy robustness under varying conditions. Furthermore, Meta-DDPG-MAML accelerates convergence rates by 30%-50% in intricate environments while achieving returns that are 20%-40% greater, thereby underscoring its efficiency and adaptability. This framework highlights the potential of integrating MAML into DRL methodologies for practical applications requiring rapid learning and robust performance. This work significantly enhances the practical applicability of DRL by enabling rapid adaptation and robust performance in real-world continuous control tasks. • Introduces Meta-DDPG-MAML, combining DDPG and MAML for rapid adaptation in continuous control tasks. • Achieves up to 50% faster convergence and 20%-40% higher returns compared to DDPG in benchmark environments. • Demonstrates improved performance across LunarLander, BipedalWalker, and Pendulum environments.
Authentication schemes for Internet of Things (IoT) networks: A systematic review and security assessment Jameel Shehu Yalli, Mohd Hilmi Hasan, Low Tan Jung, Safwan Mahmood Al-Selwi Internet of Things the Netherlands, 2025 Internet of Things (IoT) networks are revolutionizing various aspects of our lives by enabling intelligent and automated systems. However, the proliferation of IoT devices also presents significant security challenges, particularly in terms of privacy and information security. Authentication, a fundamental security mechanism, is crucial to protecting IoT devices and the sensitive data they exchange. This work provides a comprehensive review of authentication schemes for IoT devices, analyzing their strengths, weaknesses, and security considerations. We delve into the formal and informal security evaluations of these schemes, as well as their performance metrics in terms of computational overhead, communication cost, and energy consumption. By comparing/contrasting different authentication approaches, we identify potential areas for improvement and highlight the need for robust, efficient, and secure authentication solutions tailored to the unique requirements of IoT networks. This review aims to guide IoT researchers and practitioners in developing innovative authentication schemes, particularly for resource-constrained IoT devices, that can address the evolving security landscape of IoT.
Gradient Penalty Sine Cosine Algorithm (GP-SCA) Alawi Alqushaibi, Mohd Hilmi Hasan, Said Jadid Abdulkadir, Kamaluddeen Usman Danyaro, Safwan Mahmood Al-Selwi, Mohammed Gamal Ragab, Ebrahim Hamid Sumiea Lecture Notes in Electrical Engineering, 2025
A Systematic Review for Evaluating IoT Security: A Focus on Authentication, Protocols and Enabling Technologies Jameel S. Yalli, Mohd H. Hasan, Low T. Jung, Abdulrasheed I. Yerima, Dahiru A. Aliyu, Umar D. Maiwada, Safwan M. Al-Selwi, Mujeeb U. R. Shaikh IEEE Internet of Things Journal, 2025 The idea of the Internet of Things (IoT) is to connect objects and smart devices with any digital element seamlessly at anytime, anywhere, with anything. These devices or objects are characterized by low power consumption, lower bandwidth usage, limited processing power, and limited memory capacity. This article presents a systematic literature review (SLR) of the state-of-the-art security challenges of IoT systems designed for authentication. It reports the strengths, weaknesses, threats, and attacks associated with the proposed authentication models. In addition, a comprehensive analysis of IoT-compatible protocols, enabling technologies, and countermeasures to mitigate risk in IoT is investigated. This study uses the PRISMA methodology to review peer-reviewed articles published from 2014 to 2023, studying 100+ papers by selecting approximately 10 papers each year. It provides a detailed overview that contrasts and compares the advances in securing IoT devices to date while identifying remaining open research questions for IoT security designs. This article highlights the need for a survey integrating authentication models with compatible protocols and other lightweight technologies. The examination and analysis in this work contribute to the body of knowledge and serve as a roadmap for researchers and practitioners to improve their work while designing IoT security models.
A Hybrid Deep Neural Network for Early Diagnosis of Alzheimer's Disease Alawi Alqushaibi, Mohd Hilmi Hasan, Said Jadid Abdulkadir, Kamaluddeen Usman Danyaro, Rao Faizan Ali, Mohammed Gamal Ragab, Ebrahim Hamid Sumiea, Safwan Mahmood Al-Selwi Lecture Notes in Electrical Engineering, 2025
An Optimized CNN-LSTM Model for Detecting Cardiac Arrhythmias Shahab Ul Hassan, Said Jadid Abdulkadir, Mohd Soper Mohd Zahid, Abdul Muiz Fayyaz, Safwan Mahmood Al-Selwi, Ebrahim Hamid Sumiea 2024 IEEE 8th International Conference on Signal and Image Processing Applications Icsipa 2024, 2024
Mosquito Optimizer: A Nature-inspired Meta-heuristic Algorithm Alawi Alqushaibi, Mohd Hilmi Hasan, Said Jadid Abdulkadir, Kamaluddeen Usman Danyaro, Safwan Mahmood Al-Selwi, Mohammed Gamal Ragab, Ebrahim Hamid Sumiea, Abdulhadi Al-Zawari, Saipunidzam Mahamad 2024 8th International Conference on Computing Communication Control and Automation Iccubea 2024, 2024
A Comparative Analysis of Machine Learning Techniques for Heart Disease Prediction Alawi Alqushaibi, Mohammed Gamal Ragab, Ebrahim Hamid Sumiea, Safwan Mahmood Al-Selwi, Mohd Hilmi Hasan, Said Jadid Abdulkadir, Hitham Alhussian, Mahmoud Mustafa Al-Asbahi 2023 IEEE 21st Student Conference on Research and Development Scored 2023, 2023
RECENT SCHOLAR PUBLICATIONS
Explainable deep learning-based lung cancer diagnosis using clinically-guided local interpretable model-agnostic explanations SU Hassan, SJ Abdulkadir, HS Alhussian, AM Fayyaz, SM Al-Selwi, ... Scientific Reports , 2026 2026
A frequency-domain decomposition and TCN-GTAF fusion framework for GNSS sequence forecasting C Wei, Z Quan, Y Su, H Pang, L Wang, SM Shuhidan, SM Al-Selwi, ... Information Sciences, 123100 , 2026 2026
Remaining useful life estimation of turbofan engine: a sliding time window approach using deep learning A Alqushaibi, MH Hasan, SJ Abdulkadir, SM Taib, SM Al-Selwi, ... Indonesian Journal of Electrical Engineering and Computer Science (IJEECS … , 2025 2025
Grad-CAM (Gradient-weighted Class Activation Mapping): A systematic literature review AM Fayyaz, SJ Abdulkadir, N Talpur, SM Al-Selwi, SU Hassan, ... Computers in Biology and Medicine 198, 111200 , 2025 2025 Citations: 28
Optimized Long Short-Term Memory with an Adaptive Aquila Optimizer for long-term dependency problems SM Al-Selwi, MF Hassan, SJ Abdulkadir https://utpedia.utp.edu.my/id/eprint/32748/ , 2025 2025
Gradient Penalty Sine Cosine Algorithm (GP-SCA) A Alqushaibi, MH Hasan, SJ Abdulkadir, KU Danyaro, SM Al-Selwi, ... International Conference on Smart Cities, 219-231 , 2025 2025
A Hybrid Deep Neural Network for Early Diagnosis of Alzheimer's Disease A Alqushaibil, MH Hasan, SJ Abdulkadir, KU Danyaro, RF Ali, MG Ragab, ... Proceedings of the International Conference on Smart Cities-Volume 2: ICSC … , 2025 2025
The role of advanced machine learning in COVID-19 medical imaging: A technical review AM Fayyaz, SJ Abdulkadir, SU Hassan, SM Al-Selwi, EH Sumiea, LF Talib Results in Engineering 26, 105154 , 2025 2025 Citations: 3
Deep deterministic policy gradient-model-agnostic meta-learning framework: Efficient adaptation in continuous control tasks EH Sumiea, SJ Abdulkadir, HS Alhussian, SM Al-Selwi, A Alqushaibi, ... Results in Engineering 26, 105139 , 2025 2025 Citations: 9
MedDeepNet: A Feature-Enhanced Deep Learning Architecture with Lime-Based Interpretation for Lung Cancer Diagnosis SU Hassan, SJ Abdulkader, AM Fayyaz, SM Al-Selwi Available at SSRN 5258434 , 2025 2025 Citations: 1
A Novel CNN Model with Entropy-Coded Genetic Algorithm for Blood Cell Classification AM Fayyaz, SJ Abdulkadir, SM Al-Selwi, EH Sumiea, S Iqbal, SU Hassan Journal of Advanced Research in Applied Sciences and Engineering Technology … , 2025 2025 Citations: 3
A systematic review for evaluating IoT security: A focus on authentication, protocols and enabling technologies JS Yalli, MH Hasan, LT Jung, AI Yerima, DA Aliyu, UD Maiwada, ... IEEE Internet of Things Journal 12 (12), 18908-18928 , 2025 2025 Citations: 29
Authentication schemes for Internet of Things (IoT) networks: A systematic review and security assessment JS Yalli, MH Hasan, LT Jung, SM Al-Selwi Internet of Things 30, 101469 , 2025 2025 Citations: 34
Brain Tumor Detection Instance Segmentation Using YOLOv8 in Medical Imaging MG Ragab, KU Danyaro, SJ Abdulkadir, SM Al-Selwi, EH Sumiea, ... 2024 IEEE Asia-Pacific Conference on Applied Electromagnetics (APACE), 55-58 , 2024 2024 Citations: 3
Local interpretable model-agnostic explanation approach for medical imaging analysis: a systematic literature review SU Hassan, SJ Abdulkadir, MSM Zahid, SM Al-Selwi Computers in biology and medicine 185, 109569 , 2024 2024 Citations: 102
Smart grid stability prediction using adaptive aquila optimizer and ensemble stacked bilstm SM Al-Selwi, MF Hassan, SJ Abdulkadir, MG Ragab, A Alqushaibi, ... Results in Engineering 24, 103261 , 2024 2024 Citations: 21
An optimized CNN-LSTM model for detecting cardiac arrhythmias SU Hassan, SJ Abdulkadir, MSM Zahid, AM Fayyaz, SM Al-Selwi, ... 2024 IEEE 8th International Conference on Signal and Image Processing … , 2024 2024 Citations: 7
A novel deep CNN model with entropy coded sine cosine for corn disease classification MM Malik, AM Fayyaz, M Yasmin, SJ Abdulkadir, SM Al-Selwi, M Raza, ... Journal of King Saud University-Computer and Information Sciences 36 (7), 102126 , 2024 2024 Citations: 33
Mosquito Optimizer: A Nature-inspired Meta-heuristic Algorithm A Alqushaibi, MH Hasan, SJ Abdulkadir, KU Danyaro, SM Al-Selwi, ... 2024 8th International Conference on Computing, Communication, Control and … , 2024 2024
RNN-LSTM: From applications to modeling techniques and beyond—Systematic review SM Al-Selwi, MF Hassan, SJ Abdulkadir, A Muneer, EH Sumiea, ... Journal of King Saud University-Computer and Information Sciences 36 (5), 102068 , 2024 2024 Citations: 606
MOST CITED SCHOLAR PUBLICATIONS
RNN-LSTM: From applications to modeling techniques and beyond—Systematic review SM Al-Selwi, MF Hassan, SJ Abdulkadir, A Muneer, EH Sumiea, ... Journal of King Saud University-Computer and Information Sciences 36 (5), 102068 , 2024 2024 Citations: 606
A Comprehensive Systematic Review of YOLO for Medical Object Detection (2018 to 2023) MG Ragab, SJ Abdulkader, A Muneer, A Alqushaibi, EH Sumiea, ... IEEE Access 12, 57815 - 57836 , 2024 2024 Citations: 427
Deep deterministic policy gradient algorithm: A systematic review EH Sumiea, SJ Abdulkadir, HS Alhussian, SM Al-Selwi, A Alqushaibi, ... Heliyon 10 (9) , 2024 2024 Citations: 235
LSTM inefficiency in long-term dependencies regression problems SM Al-Selwi, MF Hassan, SJ Abdulkadir, A Muneer Journal of Advanced Research in Applied Sciences and Engineering Technology … , 2023 2023 Citations: 187
Local interpretable model-agnostic explanation approach for medical imaging analysis: a systematic literature review SU Hassan, SJ Abdulkadir, MSM Zahid, SM Al-Selwi Computers in biology and medicine 185, 109569 , 2024 2024 Citations: 102
Authentication schemes for Internet of Things (IoT) networks: A systematic review and security assessment JS Yalli, MH Hasan, LT Jung, SM Al-Selwi Internet of Things 30, 101469 , 2025 2025 Citations: 34
A novel deep CNN model with entropy coded sine cosine for corn disease classification MM Malik, AM Fayyaz, M Yasmin, SJ Abdulkadir, SM Al-Selwi, M Raza, ... Journal of King Saud University-Computer and Information Sciences 36 (7), 102126 , 2024 2024 Citations: 33
A systematic review for evaluating IoT security: A focus on authentication, protocols and enabling technologies JS Yalli, MH Hasan, LT Jung, AI Yerima, DA Aliyu, UD Maiwada, ... IEEE Internet of Things Journal 12 (12), 18908-18928 , 2025 2025 Citations: 29
Grad-CAM (Gradient-weighted Class Activation Mapping): A systematic literature review AM Fayyaz, SJ Abdulkadir, N Talpur, SM Al-Selwi, SU Hassan, ... Computers in Biology and Medicine 198, 111200 , 2025 2025 Citations: 28
Enhanced deep deterministic policy gradient algorithm using grey wolf optimizer for continuous control tasks EHH Sumiea, SJ Abdulkadir, MG Ragab, SM Al-Selwi, SM Fati, ... IEEE Access 11, 139771-139784 , 2023 2023 Citations: 24
Smart grid stability prediction using adaptive aquila optimizer and ensemble stacked bilstm SM Al-Selwi, MF Hassan, SJ Abdulkadir, MG Ragab, A Alqushaibi, ... Results in Engineering 24, 103261 , 2024 2024 Citations: 21
Enhanced colon cancer segmentation and image synthesis through advanced generative adversarial networks based-sine cosine algorithm A Alqushaibi, MH Hasan, SJ Abdulkadir, KU Danyaro, MG Ragab, ... IEEE Access 12, 105354-105369 , 2024 2024 Citations: 17
Deep deterministic policy gradient-model-agnostic meta-learning framework: Efficient adaptation in continuous control tasks EH Sumiea, SJ Abdulkadir, HS Alhussian, SM Al-Selwi, A Alqushaibi, ... Results in Engineering 26, 105139 , 2025 2025 Citations: 9
An optimized CNN-LSTM model for detecting cardiac arrhythmias SU Hassan, SJ Abdulkadir, MSM Zahid, AM Fayyaz, SM Al-Selwi, ... 2024 IEEE 8th International Conference on Signal and Image Processing … , 2024 2024 Citations: 7
Exploration decay policy (edp) to enhanced exploration-exploitation trade-off in ddpg for continuous action control optimization EH Sumiea, SJ AbdulKadir, H Alhussian, SM Al-Selwi, MG Ragab, ... 2023 IEEE 21st Student Conference on Research and Development (SCOReD), 19-26 , 2023 2023 Citations: 5
The role of advanced machine learning in COVID-19 medical imaging: A technical review AM Fayyaz, SJ Abdulkadir, SU Hassan, SM Al-Selwi, EH Sumiea, LF Talib Results in Engineering 26, 105154 , 2025 2025 Citations: 3
A Novel CNN Model with Entropy-Coded Genetic Algorithm for Blood Cell Classification AM Fayyaz, SJ Abdulkadir, SM Al-Selwi, EH Sumiea, S Iqbal, SU Hassan Journal of Advanced Research in Applied Sciences and Engineering Technology … , 2025 2025 Citations: 3
Brain Tumor Detection Instance Segmentation Using YOLOv8 in Medical Imaging MG Ragab, KU Danyaro, SJ Abdulkadir, SM Al-Selwi, EH Sumiea, ... 2024 IEEE Asia-Pacific Conference on Applied Electromagnetics (APACE), 55-58 , 2024 2024 Citations: 3
A comparative analysis of machine learning techniques for heart disease prediction A Alqushaibi, MG Ragab, EH Sumiea, SM Al-Selwi, MH Hasan, ... 2023 IEEE 21st Student Conference on Research and Development (SCOReD), 579-584 , 2023 2023 Citations: 3
MedDeepNet: A Feature-Enhanced Deep Learning Architecture with Lime-Based Interpretation for Lung Cancer Diagnosis SU Hassan, SJ Abdulkader, AM Fayyaz, SM Al-Selwi Available at SSRN 5258434 , 2025 2025 Citations: 1
