Abdulmajeed Alqhatani

Diagnosing lung cancer remains challenging because of the subtle differences between benign versus malignant disease on imaging studies. While traditional, convolutional neural networks (CNNs) have been the mainstay of medical imaging analysis, they still have limited applicability and robustness with variation of imaging methods. The advancement of machine learning is transforming traditional methods of diagnosis in consumer-facing healthcare technologies, changing processes in accessibility and personalized patient care. This paper describes the use of Vision Transformers (ViTs) for the task of lung cancer classification with CT scans, associated with consumer health applications. The Vision Transformer model was trained on complete dataset of 1,190 CT images, and achieved a classification rate of 98.18% with a Matthews Correlation Coefficient of 0.9676. Additionally, the model's performance was validated on simulated real-time scenarios using real-time noises and blurred datasets. Although preserving validation methods with strong diagnostic accuracy, across the entire dataset the ViT model also outperformed the conventional validated method by showing an accuracy of between 80-88% to differentiate between the noisy images and fuzzy images, even under these circumstances. While initial results provide promising information about the robustness of ViTs when dealing with image variation, a cautious approach should still be exercised in contextualizing results since the evaluation studies were completed within a relatively small dataset and simulation settings. It will be valuable for future research studies to use larger datasets that are more representative of true clinical conditions and datasets that are clinically validated to determine whether ViTs are advantageous to clinical identification of oncology diagnostic and to improve early detection.

Diagnosing lung cancer remains challenging because of the subtle differences between benign versus malignant disease on imaging studies. While traditional, convolutional neural networks (CNNs) have been the mainstay of medical imaging analysis, they still have limited applicability and robustness with variation of imaging methods. The advancement of machine learning is transforming traditional methods of diagnosis in consumer-facing healthcare technologies, changing processes in accessibility and personalized patient care. This paper describes the use of Vision Transformers (ViTs) for the task of lung cancer classification with CT scans, associated with consumer health applications. The Vision Transformer model was trained on complete dataset of 1,190 CT images, and achieved a classification rate of 98.18% with a Matthews Correlation Coefficient of 0.9676. Additionally, the model's performance was validated on simulated real-time scenarios using real-time noises and blurred datasets. Although preserving validation methods with strong diagnostic accuracy, across the entire dataset the ViT model also outperformed the conventional validated method by showing an accuracy of between 80-88% to differentiate between the noisy images and fuzzy images, even under these circumstances. While initial results provide promising information about the robustness of ViTs when dealing with image variation, a cautious approach should still be exercised in contextualizing results since the evaluation studies were completed within a relatively small dataset and simulation settings. It will be valuable for future research studies to use larger datasets that are more representative of true clinical conditions and datasets that are clinically validated to determine whether ViTs are advantageous to clinical identification of oncology diagnostic and to improve early detection.

Unmanned aerial vehicles (UAVs) have enabled numerous inventive solutions to multiple problems, considerably facilitating our daily lives; however, UAVs frequently rely on an open wireless channel for communication, making them susceptible to cyber-physical threats. Also, UAVs cannot execute complicated cryptographic algorithms due to their limited onboard computing capabilities. Balancing high-security levels and minimum computation costs is imperative when developing a security solution for UAVs. Consequently, several proxy signature schemes have been proposed in the literature to fulfill these requirements. Nevertheless, many of these solutions face the issue of high computation costs, and some exhibit security vulnerabilities that could not be more feasible options for UAV communication. Considering these constraints in mind, in this article, we introduce an improvised certificate-based proxy signature scheme (ICPS), which leverages the concept of hyperelliptic curve cryptography (HECC) to meet the security and efficiency requirements of UAV networks. The proposed ICPS scheme offers a range of notable features, including its ability to address key escrow and secret key distribution issues. The proposed ICPS scheme’s security hardness has been evaluated using the widely known security tool, the random oracle model (ROM), proving its resilience against known and unknown cybersecurity threats. Finally, this study conducts a performance comparison of the proposed scheme against existing schemes, emphasizing its outstanding cost-efficiency. Notably, the computation cost is measured at 5.3536 ms and the communication cost at 1120 bits, substantially lower than relevant existing schemes.

Recommender systems are already being used by several biggest e-commerce websites to assist users in finding things to buy. A recommender system gains knowledge from a consumer and suggests goods from the available goods that will find most value. In this deep learning technique, the Hybrid Deep Collaborative Transformer (HDCT) method has emerged as a promising approach. However, it is crucial to thoroughly examine and rectify any potential errors or limitations in the optimization process to ensure the optimal performance of the HDCT model. This study aims to address this concern by thoroughly evaluating the HDCT method uncovering any underlying errors or shortcomings. By comparing its performance against other existing models, the proposed HDCT with Federated Learning method demonstrates superior recommendation accuracy and effectiveness. Through a comprehensive analysis, this research identifies and rectifies the errors in the HDCT model, thereby enhancing its overall performance. The findings of this study provide valuable insights for researchers and practitioners in the field of e-commerce recommendation systems. Data for the RS is collected from the Myntra fashion product dataset. By understanding and addressing the limitations of the HDCT method, businesses can leverage its advantages to improve customer satisfaction and boost their revenue. Ultimately, this research contributes to the ongoing advancements in e-commerce recommendation systems and paves the way for future improvements in this rapidly evolving domain. The suggested model’s efficacy is assessed using metrics for MSE, MSRE, NMSE, RMSE, and MAPE. The suggested values in metrics are 0.2971, 0.2763, 0.4013, 0.3222, 0.2911 at a 70% learn rate and 0.2403, 0.2234, 0.3506, 0.2025, 0.2597 at an 80% learn rate, and the proposed model outperformed with the least amount of error.

The Internet of vehicles (IoVs) is an innovative paradigm which ensures a safe journey by communicating with other vehicles. It involves a basic safety message (BSM) that contains sensitive information in a plain text that can be subverted by an adversary. To reduce such attacks, a pool of pseudonyms is allotted which are changed regularly in different zones or contexts. In base schemes, the BSM is sent to neighbors just by considering their speed. However, this parameter is not enough because network topology is very dynamic and vehicles can change their route at any time. This problem increases pseudonym consumption which ultimately increases communication overhead, increases traceability and has high BSM loss. This paper presents an efficient pseudonym consumption protocol (EPCP) which considers the vehicles in the same direction, and similar estimated location. The BSM is shared only to these relevant vehicles. The performance of the purposed scheme in contrast to base schemes is validated via extensive simulations. The results prove that the proposed EPCP technique outperformed compared to its counterparts in terms of pseudonym consumption, BSM loss rate and achieved traceability.

Cybersecurity continues to be a major issue for all industries engaged in digital activity given the cyclical surge in security incidents. Since more Internet of Things (IoT) devices are being used in homes, offices, transportation, healthcare, and other venues, malicious attacks are happening more frequently. Since distance between IoT as well as fog devices is closer than distance between IoT devices as well as the cloud, attacks can be quickly detected by integrating fog computing into IoT. Due to the vast amount of data produced by IoT devices, ML is commonly employed for attack detection. This research proposes novel technique in cybersecurity-based network traffic analysis and malicious attack detection using IoT artificial intelligence techniques for a sustainable smart city. A traffic analysis has been carried out using a kernel quadratic vector discriminant machine which enhances the data transmission by reducing network traffic. This enhances energy efficiency with reduced traffic. Then, the malicious attack detection is carried out using adversarial Bayesian belief networks. The experimental analysis has been carried out in terms of throughput, data traffic analysis, end-end delay, packet delivery ratio, energy efficiency, and QoS. The proposed technique attained a throughput of 98%, data traffic analysis of 74%, end-end delay of 45%, packet delivery ratio of 92%, energy efficiency of 92%, and QoS of 79%.

The widespread usage of smart devices with various city-centric services speeds up and improves civic life, in contrast to growing privacy and security concerns. Security issues are exacerbated when e-government service providers trade their services within a centralised framework. Due to security concerns, city-centric centralised services are being converted to blockchain-based systems, which is a very time-consuming and challenging process. The interoperability of these blockchain-based systems is also more challenging due to protocol variances, an excessive amount of local transactions that raise scalability and rapidly occupy memory. In this paper, we have proposed a framework for interoperability across various blockchain-based smart city services. It also summarises how independent service providers might continue self-service choices (i.e., local transactions) without overloading the blockchain network and other organisations. A simulated interoperability network is used to show the network’s effectiveness. The experimental outcomes show the scalability and memory optimization of the blockchain network.

Early detection of abnormalities in chest X-rays is essential for COVID-19 diagnosis and analysis. It can be effective for controlling pandemic spread by contact tracing, as well as for effective treatment of COVID-19 infection. In the proposed work, we presented a deep hybrid learning-based framework for the detection of COVID-19 using chest X-ray images. We developed a novel computationally light and optimized deep Convolutional Neural Networks (CNNs) based framework for chest X-ray analysis. We proposed a new COV-Net to learn COVID-specific patterns from chest X-rays and employed several machine learning classifiers to enhance the discrimination power of the presented framework. Systematic exploitation of max-pooling operations facilitates the proposed COV-Net in learning the boundaries of infected patterns in chest X-rays and helps for multi-class classification of two diverse infection types along with normal images. The proposed framework has been evaluated on a publicly available benchmark dataset containing X-ray images of coronavirus-infected, pneumonia-infected, and normal patients. The empirical performance of the proposed method with developed COV-Net and support vector machine is compared with the state-of-the-art deep models which show that the proposed deep hybrid learning-based method achieves 96.69% recall, 96.72% precision, 96.73% accuracy, and 96.71% F-score. For multi-class classification and binary classification of COVID-19 and pneumonia, the proposed model achieved 99.21% recall, 99.22% precision, 99.21% F-score, and 99.23% accuracy.