Ahmed S. Alfakeeh

Skin cancer is among the most aggressive and prevalent forms of cancer worldwide, with melanoma posing a high risk of metastasis and mortality when not detected early. Manual diagnosis by dermatologists, while effective, faces challenges such as subjectivity, variability, and limited accessibility in underserved regions. To address these limitations, this study proposes a robust computer-aided diagnostic system for early detection of skin cancer using a hybrid feature extraction approach and machine learning. In this study, several methodologies were developed for the automated classification of dermoscopic images, with a primary focus on a hybrid diagnostic model combining radiomic and deep learning features. Specifically, a Random Forest (RF) classifier was trained on fused features extracted from radiomic algorithms and deep convolutional layers of CNN. The proposed MobileNetV2 + radiomics + RF model achieved outstanding performance, with an average AUC of 85.09%, accuracy of 94.7%, sensitivity of 84.53% and specificity exceeding 99.2%. It exhibited particularly strong classification capabilities for high-risk lesions such as melanoma (AUC of 94.4%) and nevi (AUC of 98.2%), while maintaining robust performance across other lesion types. The integration of radiomic and CNN-based features through an RF classifier offers a highly effective approach for early skin cancer detection, with significant implications for clinical practice.

Melanoma presents considerable risks as it is an extremely aggressive type of skin cancer that rapidly metastasize if not addressed promptly. Early detection of melanoma significantly improves treatment outcomes and reduces mortality, aligning with SDG 3 (Good Health and Well-being). Dermatologists face challenges in accurately diagnosing skin lesions and recognizing skin cancer early due to human error and variation in experience. This study develops hybrid Artificial Neural Network (ANN) methodologies integrating CNN-based feature selection algorithms for dermoscopic image analysis, contributing to health system sustainability by enhancing diagnostic accuracy through AI. This approach, tested on the ISIC-2019 dataset, employs advanced filtering and segmentation along with fused multi-model CNN features and Ant Colony Optimization (ACO) for feature selection. The fused features generated from MobileNet, ResNet34, and GoogLeNet are subsequently processed by an Artificial Neural Network that acts as the final classifier, enabling the practical deployment of the model as an automated decision-support tool for early melanoma screening in clinical and tele-dermatology environments. The ANN that integrates features from MobileNet + ResNet34 + GoogLeNet models achieved an AUC (94.74%), sensitivity (94.61%), accuracy (98.2%), precision (95.24%), and specificity (99.61%). This result supporting SDG targets related to reducing premature mortality from non-communicable diseases and promoting sustainable health systems. Compared with previously reported single-CNN and hybrid models (accuracy range 81–95%), the proposed framework exhibits a clear improvement of approximately 3–7% in diagnostic accuracy and enhanced stability across minority lesion classes, underscoring its reliability for early melanoma detection and its alignment with SDG targets aimed at reducing premature mortality from non-communicable diseases.

Location-based services (LBSs) are a crucial element of the Internet of Things (IoT) and have garnered significant attention from both researchers and users, driven by the rise of wireless devices and a growing user base. However, the use of LBS-enabled applications carries several risks, as users must provide their real locations with each query. This can expose them to potential attacks from the LBS server, leading to serious issues like the theft of personal information. Consequently, protecting location privacy is a vital concern. To address this, location dummy-based methods are employed to safeguard the location privacy of LBS users. However, location dummy-based approaches also suffer from problems such as low resistance against inference attacks and the generation of strong dummy locations, an issue that is considered an open problem. Moreover, generating many location dummies to achieve a high privacy protection level leads to high network overhead and requires high computational capabilities on the mobile devices of the LBS users, and such devices are limited. In this paper, we introduce the Caching-Aware Double-Dummy Selection (CaDDSL) algorithm to protect the location privacy of LBS users against homogeneity location and semantic location inference attacks, which may be applied by the LBS server as a malicious party. Then, we enhance the CaDDSL algorithm via encapsulation with agents to solve the tradeoff between generating many dummies and large network overhead by proposing the Cache-Aware Overhead-Aware Dummy Selection (CaOaDSL) algorithm. Compared to three well-known approaches, namely GridDummy, CirDummy, and Dest-Ex, our approach showed better performance in terms of communication cost, cache hit ratio, resistance against inference attacks, and network overhead.

Industrial healthcare systems aim to provide progressive real-time monitoring of patients and improve medical services through data sharing among intelligent wearable sensors and devices. However, this connectivity carries fundamental vulnerabilities associated with privacy and security because of the requirement of constant communication and monitoring over a public network. However, new technologies, namely Deep Learning (DL) and Blockchain (BC), are promising to overcome these problems and modernize healthcare systems. BC technology is receiving important attention because of its permanent and decentralized nature. Incorporating the immutable and decentralized nature of BC technology with Intrusion Detection Systems (IDS) can provide a stronger and more trustworthy security structure for IoT healthcare systems. This study presents a Secure Internet of Things Framework Using Blockchain and Advanced Feature Selection (SIoTF-BCAFS) model for smart healthcare. The aim is to improve security and reliability in IoT-enabled smart healthcare systems using advanced techniques. Initially, BC-assisted data transmission is employed to ensure secure and transparent communication between devices, especially in IoT. Then, min-max scaling is applied in the data preprocessing step to convert the input data. The ReliefF technique is used to select the best features. Additionally, the SIoTF-BCAFS technique implements an Autoencoder (AE) along with a Temporal Convolutional Network (TCN) model to effectively capture sequential patterns and temporal dependencies. The experimental analysis of the SIoTF-BCAFS technique was performed on the ToN-IoT dataset, demonstrating superior accuracy over existing approaches.

Recently, the Internet of Things (IoT) has attracted the attention of many researchers due to its popularity in various applications. Connecting different capability devices in the Internet of Things architecture makes security a big challenge. Datagram Transport Layer Security Protocol is considered a standard for securing communication among Internet of Things devices by establishing a secret key. However, the default method, X.509, requires certificates and public key infrastructure that are more resource-consuming and not suitable for resource-constrained devices. On the other hand, datagram transport layer security supports the pre-shared key approach and raw pubic keys, which are lightweight but not scalable for such large networks. Hence, a scalable and lightweight mutual key establishment and management protocol is proposed for such a large number of resource-constrained IoT devices. The implementation of the proposed scheme in Contiki OS and on a real IoT platform shows its performance evaluation in terms of feasibility and scalability.

Wireless multimedia sensor networks (WMSNs) have stringent constraints and need to deliver data packets to the sink node within a predefined limited time. However, due to congestion, buffer overflow occurs and leads to the degradation of the quality-of-service (QoS) parameters of event information. Congestion in WMSNs results in exhausted node energy, degraded network performance, increased transmission delays, and high packet loss. Congestion occurs when the volume of data trying to pass through a network exceeds its capacity. First, the BOCC protocol uses two congestion indicators to detect congestion. One is the buffer occupancy and other is the buffer occupancy change rate. Second, a rate controller is proposed to protect high-priority I-frame packets during congestion. BOCC sends a congestion notification to the source node to reduce congestion in the network. The source node adjusts its data transmission rate after receiving the congestion notification message. In the proposed algorithm, the rate adjustment is made by discarding low-priority P-frame packets from the source nodes. Third, to further improve the performance of the BOCC protocol, the problem is formulated as a constrained optimization problem and solved using convex optimization and sequential quadratic programming (SQP) methods. Experimental results based on Raspberry Pi sensor nodes show that the BOCC protocol achieves up to 16% reduction in packet loss and up to 23% reduction in average end-to-end delay compared to state-of-the-art congestion control algorithms.

Fog computing is a key component of future intelligent transportation systems (ITSs) that can support the high computation and large storage requirements needed for autonomous driving applications. A major challenge in such fog-enabled ITS networks is the design of algorithms that can reduce the computation times of different tasks by efficiently utilizing available computational resources. In this paper, we propose a data-enabled cooperative technique that offloads some parts of a task to the nearest fog roadside unit (RSU), depending on the current channel quality indicator (CQI). The rest of the task is offloaded to a nearby cooperative computing vehicle with available computing resources. We developed a cooperative computing vehicle selection technique using an artificial neural network (ANN)-based prediction model that predicts both the computing availability once the task is offloaded to the potential computing vehicle and the link connectivity when the task result is to be transmitted back to the source vehicle. Using detailed simulation results in MATLAB 2020a software, we show the accuracy of our proposed prediction model. Furthermore, we also show that the proposed technique reduces total task delay by 37% compared to other techniques reported in the literature.

Antibiotic-resistant bacteria have proliferated at an alarming rate as a result of the extensive use of antibiotics and the paucity of new medication research. The possibility that an antibiotic-resistant bacterial infection would progress to sepsis is one of the major collateral problems affecting people with this condition. 31,000 lives were lost due to sepsis in England with costs about two billion pounds annually. This research aims to develop and evaluate several classification approaches to improve predicting sepsis and reduce the tendency of underdiagnosis in computer-aided predictive tools. This research employs medical datasets for patients diagnosed with sepsis, and it analyses the efficacy of ensemble machine learning techniques compared to nonensemble machine learning techniques and the significance of data balancing and conditional tabular generative adversarial nets for data augmentation in producing reliable diagnosis. The average F Score obtained by the nonensemble models trained in this paper is 0.83 compared to the ensemble techniques average of 0.94. Nonensemble techniques, such as Decision Tree, achieved an F score of 0.90, an AUC of 0.90, and an accuracy of 90. Histogram-basedgradient boosting classification tree achieved an F score of 0.96, an AUC of 0.96, and an accuracy of 95, surpassing the other models tested. Additionally, when compared to the current state-of-the-art sepsis prediction models, the models developed in this study demonstrated higher average performance in all metrics, indicating reduced bias and improved robustness through data balancing and conditional tabular generative adversarial nets for data augmentation. The study revealed that data balancing and augmentation on the ensemble machine learning algorithms boost the efficacy of clinical predictive models and can help clinics decide which data types are most important when examining patients and diagnosing sepsis early through intelligent human-machine interface.