Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Nashwan Adnan Othman has a total of 71 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Task Scheduling in Cloud Computing Using a Hybrid Bobcat Optimization Algorithm with Humboldt Squid Optimization

    by Vallu Visrutatma Rao, Winner Pulakhandam, Archana Chaluvadi, N Purandhar, Qaisar Abbas, Nashwan Adnan Othman

    Task scheduling is the most significant resource optimization and efficient system performance method employed in cloud computing. This paper presents the Hybrid Bobcat Optimization Algorithm (BSHO), a novel method that hybridizes the Bobcat Optimization Algorithm (BOA) and Humboldt Squid Optimization (HSO) to reduce the complexity of task scheduling in a dynamic cloud environment. The goal of the suggested framework is to maximize task allocation to virtual machines (VMs), minimize makespan, optimize the usage of resources, and minimize energy consumption. BSHO leverages the local search feature of BOA and the global search techniques of HSO and thus facilitates adaptability to diverse cloud environments and workloads. The performance of BSHO is evaluated using significant parameters such as execution cost, task completion rate, CPU utilization, and fitness value. Experiment outcomes indicate that BSHO is better than classic scheduling algorithms in that it achieves a lower cost of execution ranging from $800 to $1300, while reducing the task completion time to between 20 ms and 45[ ]ms based on the VM number. In addition, BSHO ensures a maximum load balancing effectiveness of up to 89% and a rate of task execution of 98%, thus ranking as a highly effective scheduling algorithm in cloud computing systems. The outcomes warrant that BSHO presents a solid, efficient, and scalable solution for improving cloud task scheduling.


  • Optimized Acute Kidney Injury Monitoring: Integrating Long Short-Term Memory (LSTM) Networks with Ant Colony Optimization (ACO)

    by Archana Chaluvadi, Nashwan Adnan Othman, Winner Pulakhandam, Visrutatma Rao Vallu, Revathi S

    Acute Kidney Injury (AKI) is a serious Intensive Care Unit (ICU) complication in high-risk patients with resultant increased morbidity and mortality. Early detection and intervention in AKI are important to enhance patient outcomes. This paper presents a predictive framework that integrates Long Short-Term Memory (LSTM) networks with Ant Colony Optimization (ACO) to predict AKI from patient data in the MIMIC-IV database. We had 34,618 critically ill patients in our cohort, with 48% developing in-hospital AKI. The framework employs ACO-based feature selection to identify significant clinical variables including renal biomarkers, electrolyte imbalance, and hemodynamic variables to enhance interpretability and computational efficiency. LSTM is trained to identify temporal relations between patient data to facilitate early and accurate prediction of AKI occurrence. Stratified data splitting was employed for unbiased evaluation to ensure robust assessment. The experiment results indicate that LSTM-ACO performs superior to conventional machine learning and deep learning algorithms including GBM, GLM, KNN, EOAEDL-CKDD, OANN and OLSTM with AUC-ROC value of 0.99, sensitivity of 95.5%, and specificity of 97.3%. ACO-based feature selection was able to reduce the dimensionality by 40% with maximum utilization of computational resources without any compromise on predictive capability. The findings point toward the potential of the LSTM-ACO model to be used in real-time clinical decision-making for the prediction of AKI in ICU settings.

  • Enhanced phishing detection framework with tokenization WOA optimization domain features and CNN classification

    by Zardasht Abdulaziz Abdulkarim Shwany, Younus Ameen Muhammed, Bishar Rasheed Ibrahim, Omar Sedqi Kareem, Mehmet Umut Salur, Nashwan Adnan Othman

    This study investigates the efficiency of numerous phishing detection techniques, concentrating on the proposed Convolutional Neural Network-based model that demonstrates superior performance over traditional and advanced methods. Attaining an exceptional 99.3% of accuracy, 99.93% of precision, 98.70% of recall, and 99.31% of F1 score, the typical leverages advanced methodologies such as the Whale Optimization Algorithm for feature selection and Zero Trust Verification to enhance robustness and efficiency. By detecting subtle patterns and differences, the CNN-based model outperforms traditional approaches like Decision Trees and KNN, as well as advanced models such as LSTM, making it a really reliable resolution for phishing detection. The model is further strengthened by its ability to integrate comprehensive feature analysis, adaptive learning, and dynamic classification capabilities. Proposed future enhancements include expanding datasets to encompass multilingual and diverse samples, incorporating hybrid models like transformers for improved feature extraction and optimizing the system for actual detection and deployment. These developments aim to recover the model’s scalability and applicability evolving cybersecurity environments, offering a comprehensive and reliable solution to effectively mitigate phishing threats in real-world scenarios.


  • Intrusion detection in SDN-IoT using robust GRU with attention mechanism and fuzzy weighting

    by Omar Sedqi Kareem, Younus Ameen Muhammed, Zardasht Abdulaziz Abdulkarim Shwany, Twana Saeed Mustafa, Mehmet Umut Salur, Nashwan Adnan Othman

    Dealing with the rise of complexity and increasing frequency of cyber threats in both SDN and IoT, the characteristic of intrusion detection in these platforms will be a predictable requirement to overcome the present attacks like malware, DDoS, and MITM, among others. This paper presents an all-inclusive and effective intrusion detection system that can combine advanced DL methods specifically concentrating on merging networks with LSTM and GRU. The model’s capability to identify critical information from network traffic data and use fuzzy weighting algorithms and attention mechanisms allows for correctly identifying anomalies. The system is qualified and authenticated on its effectiveness using the BoT-IoT dataset a standard for simulating authentic IoT traffic and attack patterns. Advanced optimization techniques like CGO and GBO improve the system’s computational efficiency by guaranteeing low resource consumption while preserving excellent performance. The proposed IDS has shown the capability to cruel system activities for a notable accuracy of 99.97, an F1 score of 99.94, and extremely low false-positive and false-negative rates. Such an adaptation has proven suitable for real-time SDN-IoT network operation within dynamic and resource-constrained systems with an outcome of effective and real-time mitigation of several types of threats in such atmospheres through smooth adaptability to any variation in nature of the traffic as well as diverse types of attack. An expressive upgrade over the conventional structures, this IDS is also a model for the future development in the security of networked devices.


  • Robust IDS for SDN-iot networks with residual GRU in two-tier PRF MAC and gaussian-gbell hybrid

    by Twana Saeed Mustafa, Omar Sedqi Kareem, Younus Ameen Muhammed, Zardasht Abdulaziz Abdulkarim Shwany, Nashwan Adnan Othman

    Network administration has been converted by the quick spread of IoT devices and the rise of SDN, but serious security risks exist. Because they frequently have limited resources, IoT devices are exposed to attacks like man-in-the-middle, brute force, and DDoS. An IDS that is strong and designed for SDN-IoT cloud networks is essential. This paper proposed a hybrid IDS framework that uses the Gaussian-Gbell and Residual GRU models to increase detection efficiency and robustness. While Residual GRU targets the long-term dependencies and vanishing gradient problems in sequential data processing, the Gaussian-Gbell hybrid, which combines Gaussian and Generalized Bell functions, improves flexibility in capturing non-linear data patterns. Using SDN-IoT traffic circumstances the proposed model shows a 99.7 percentage of accuracy rate in real-time cyber threat detection and mitigation. The architecture highlights robustness against adversarial attacks and low processing overhead for real-time activities while balancing scalability, performance and security. The model provides better F1 scores, precision, and recall than conventional methods, making it a dependable IDS solution for changing network settings. Future research aims to increase adaptation to new threats, combine federated learning for distributed IoT security and combine minimal encryption for devices with limited resources. This study emphasizes how crucial sophisticated IDS are to protecting dynamic SDN-IoT cloud networks from the ever-increasing security landscape while maintaining scalability and operational predictability.


  • Top Cited Papers
    Open Access

    Investigating the performance of AI-driven smart building systems through advanced deep learning model analysis

    by M. Arun, Debabrata Barik, Nashwan Adnan Othman, Seepana Praveenkumar, Kapura Tudu


  • Advancing Drone Control Interfaces: Implementing JavaScript for Mobile Connectivity with the Parrot AR Drone 2.0

    by Министерство Высшего Образования И Научных Исследований, Мохаммед Хусейн, Али Ихсан Алансари, Хайдер Халил Иса, Али Маджид, Университетский Колледж Аль-Калам, Ахмед Исам Аль-Хатем, Университетский Колледж Аль-Нухба, Нашван Аднан Отман, Инженерный Колледж Университет Знаний, et al.

    Проиллюстрировано использование JavaScript, языка программирования высокого уровня, для мобильного подключения к Parrot AR Drone 2.0, что делает его более доступным для более широкой аудитории. С помощью пакета AR-drone на платформе Node.js создано серверное приложение на Express.js. Эта программа преобразует HTTP-запросы GET из мобильного приложения в инструкции, которые можно выполнять с помощью дрона. Это изменение позволяет мобильному устройству управлять операциями дрона, такими как взлет, посадка и движение. С помощью этой системы дрон успешно преодолел полосу препятствий в помещении, продемонстрировав свою пригодность для решения сложных навигационных задач. JavaScript может служить мостом между технологией дронов и интерфейсами смартфонов. Эта интеграция демократизирует управление дронами и знаменует прорыв в слиянии Интернета вещей и технологий дронов. Это подчеркивает необходимость решения проблем обработки данных в реальном времени и безопасности.

  • Advancing Drone Control Interfaces: Implementing JavaScript for Mobile Connectivity with the Parrot AR Drone 2.0

    by Ministry of Higher Education and Scientific Research, Mohammed Hussein, Ali Ihsan Alanssari, Al-Nahrain University, Haider Khalil Easa, Al-Kitab University, Ali Majid, Al-Qalam University College, Ahmed Isam Al-Hatem, Al-Nukhba University College, et al.

    Background: Drones, also known as Unmanned Aerial Vehicles (UAVs), have been incorporated into current applications, shifting from military to general-purpose usage. The proliferation of hobbyist drones, such as the Parrot AR Drone 2.0, highlights the need for straightforward control systems, particularly considering the enormous number of smartphone users worldwide. Objective: This article aims to illustrate the usage of JavaScript, a high-level programming language, for mobile connectivity to the Parrot AR Drone 2.0, making it more accessible to a broader audience. Methods: Using the AR-drone package on the Node.js platform, a server-side application in Express.js was created. This program converts HTTP GET queries from a mobile app into drone-actionable instructions. This change enables a mobile device to control the drone's operations, such as takeoff, landing, and movement. Results: With this system in place, the drone successfully navigated an indoor obstacle course, demonstrating its suitability for complicated navigational tasks. The method streamlines drone operations, improving the user experience. Conclusion: JavaScript can act as a bridge between drone technology and smartphone interfaces. This integration democratizes drone control and heralds breakthroughs in the confluence of IoT and drone technology. It underscores the need to address real-time processing and security issues in future research.

  • Open Access

    Analysis of Smart Lung Tumour Detector and Stage Classifier Using Deep Learning Techniques with Internet of Things

    by Shubham Joshi, Shraddha Viraj Pandit, Piyush Kumar Shukla, Atiah H. Almalki, Nashwan Adnan Othman, Adnan Alharbi, Musah Alhassan

    The use of artificial intelligence (AI) and the Internet of Things (IoT), which is a developing technology in medical applications that assists physicians in making more informed decisions regarding patients’ courses of treatment, has become increasingly widespread in recent years in the field of healthcare. On the other hand, the number of PET scans that are being performed is rising, and radiologists are getting significantly overworked as a result. As a direct result of this, a novel approach that goes by the name “computer-aided diagnostics” is now being investigated as a potential method for reducing the tremendous workloads. A Smart Lung Tumor Detector and Stage Classifier (SLD-SC) is presented in this study as a hybrid technique for PET scans. This detector can identify the stage of a lung tumour. Following the development of the modified LSTM for the detection of lung tumours, the proposed SLD-SC went on to develop a Multilayer Convolutional Neural Network (M-CNN) for the classification of the various stages of lung cancer. This network was then modelled and validated utilising standard benchmark images. The suggested SLD-SC is now being evaluated on lung cancer pictures taken from patients with the disease. We observed that our recommended method gave good results when compared to other tactics that are currently being used in the literature. These findings were outstanding in terms of the performance metrics accuracy, recall, and precision that were assessed. As can be shown by the much better outcomes that were achieved with each of the test images that were used, our proposed method excels its rivals in a variety of respects. In addition to this, it achieves an average accuracy of 97 percent in the categorization of lung tumours, which is much higher than the accuracy achieved by the other approaches.