The internet of things (IoT) concept has been extremely investigated in many modern smart applications, which enable a set of sensors to either process the collected data locally or send them to the cloud for remote processing. Unfortunately, cloud datacenters are located far away from IoT devices, and consequently, the transmission of IoT data may be delayed. In this paper, we investigate fog computing, which is a new paradigm that overcomes many issues of cloud computing. More importantly, dynamic task offloading in fog computing is a challenging problem that requires an optimal decision for processing the tasks that are generated in each time slot. Thus, exact optimization methods based on Lyapunov function have been widely used for solving dynamic offloading which represents an NP hard problem. To overcome the scalability issue of exact optimization techniques, we have explored famous population based meta-heuristics for optimizing the offloading process of a set of dynamic tasks using Orthogonal Frequency Division Multiplexing (OFDM) communication. Hence, a parallel multi-threading framework is proposed for generating the optimal offloading solution while selecting the best sub-carrier for each offloaded task. More importantly, our contribution associates a thread for each IoT device and generates a population of random solutions. Next, each population is updated and evaluated according to the proposed fitness function that considers a tradeoff between the delay and energy consumption. Upon the arrival of new tasks at each time slot, an evaluation is performed for maintaining some individuals of the previous population while generating new individuals based on some criteria. Our results have been compared to the results achieved using Lyapunov optimization. They demonstrate the convergence of the fitness function, the scalability of the parallel Particle Swarm Optimization (PSO) approach, and the performance in terms of the offline error and the execution cost.

Vehicular Edge-Cloud Computing (VECC) paradigm has emerged as a viable approach to overcome the inherent resource limitations of vehicles by offloading computationally demanding tasks to remote servers. Despite its potential, existing offloading strategies often result in increased latency and sub-optimal performance due to the concentration of workloads on a limited number of connected Roadside Units (RSUs). Moreover, ensuring data security and optimizing energy usage within this framework remain significant challenges. To address these concerns, this paper proposes a comprehensive framework for VECC systems. A novel load-balancing algorithm is proposed to effectively redistribute vehicles among RSUs, considering factors such as RSUs load, computational capacity, and data rate. Additionally, a robust security mechanism is incorporated using the Advanced Encryption Standard (AES) in conjunction with Electrocardiogram (ECG) signals as encryption keys to enhance data protection during transmission. To further improve system efficiency, a novel caching strategy is introduced, enabling edge servers to store completed tasks, which in turn reduces both latency and energy consumption. An optimization model is also proposed to minimize energy expenditure while ensuring that latency constraints are satisfied during computation offloading. Given the complexity of this problem in large-scale vehicular networks, the study formulates an equivalent reinforcement learning model and employs a deep learning algorithm to derive optimal solutions. Simulation results conclusively demonstrate that the proposed model significantly outperforms existing benchmark techniques in terms of energy savings.

Accurate detection of fractures in X-ray images is important to initiate appropriate medical treatment in time—in this study, an advanced combined attention CNN model with multiple attention mechanisms was developed to improve fracture detection by deeply representing features. Specifically, our model incorporates squeeze blocks and convolutional block attention module (CBAM) blocks to improve the model’s ability to focus on relevant features in X-ray images. Using computed tomography X-ray images, this study assesses the diagnostic efficacy of the artificial intelligence (AI) model before and after optimization and compares its performance in detecting fractures or not. The training and evaluation dataset consists of fractured and non-fractured X-rays from various anatomical locations, including the hips, knees, lumbar region, lower limb, and upper limb. This gives an extremely high training accuracy of 99.98 and a validation accuracy 96.72. The attention-based CNN thus showcases its role in medical image analysis. This aspect further complements a point we highlighted through our research to establish the role of attention in CNN architecture-based models to achieve the desired score for fracture in a medical image, allowing the model to generalize. This study represents the first steps to improve fracture detection automatically. It also brings solid support to doctors addressing the continued time to examination, which also increases accuracy in diagnosing fractures, improving patients’ outcomes significantly.

Electrocardiogram (ECG) analysis is crucial for diagnosing cardiovascular diseases (CVD), especially atrial fibrillation (AF), a prevalent cardiac rhythm abnormality. However, the variability and complexity of ECG signals make AF classification challenging, highlighting the need for more accurate and reliable methods. This paper introduces a novel deep learning (DL) architecture designed to enhance the processing, feature extraction, and analysis of ECG signals. The proposed model utilizes a hybrid framework that combines standard and dilated convolutional networks, advanced attention mechanisms, and temporal sequence learning to address the complexities of ECG data. A parallel architecture integrates a MobileViT block for efficient spatial feature extraction and an Efficient Channel Attention (ECA) module to refine feature representations. Simultaneously, a Long Short-Term Memory (LSTM) network captures temporal dependencies in the data. The concatenation of the outputs from both the MobileViT-ECA block and the LSTM network allows for the extraction of both local and global features. Fully connected layers, along with dropout regularization, ensure robust performance and effective classification. We analyzed ECG signals of three different lengths (1, 2, and 3 seconds) to investigate how segment length affects model performance. Additionally, we conducted various classification scenarios and an ablation study to simulate real-world conditions and identify key components of the architecture. Our experimental findings show that the model attains high accuracy in ECG classification and AF identification, with average metrics including accuracy of 87.80%, recall of 87.80%, F1-score of 87.45%, and specificity of 95.66%. The proposed model outperforms existing methods on the same AF dataset, demonstrating its superior performance. This method accurately classifies ECG signals and detects AF, highlighting its potential for clinical application. The results help move forward the development of automated tools for diagnosing heart diseases and create new opportunities for flexible solutions in this field.

Task offloading is a key strategy in Fog Computing (FC). The definition of resource-constrained devices no longer applies to sensors and Internet of Things (IoT) embedded system devices alone. Smart and mobile units can also be viewed as resource-constrained devices if the power, cloud applications, and data cloud are included in the set of required resources. In a cloud-fog-based architecture, a task instance running on an end device may need to be offloaded to a fog node to complete its execution. However, in a busy network, a second offloading decision is required when the fog node becomes overloaded. The possibility of offloading a task, for the second time, to a fog or a cloud node depends to a great extent on task importance, latency constraints, and required resources. This paper presents a dynamic service that determines which tasks can endure a second offloading. The task type, latency constraints, and amount of required resources are used to select the offloading destination node. This study proposes three heuristic offloading algorithms. Each algorithm targets a specific task type. An overloaded fog node can only issue one offloading request to execute one of these algorithms according to the task offloading priority. Offloading requests are sent to a Software Defined Networking (SDN) controller. The fog node and controller determine the number of offloaded tasks. Simulation results show that the average time required to select offloading nodes was improved by 33% when compared to the dynamic fog-to-fog offloading algorithm. The distribution of workload converges to a uniform distribution when offloading latency-sensitive non-urgent tasks. The lowest offloading priority is assigned to latency-sensitive tasks with hard deadlines. At least 70% of these tasks are offloaded to fog nodes that are one to three hops away from the overloaded node.

The growing demands of Industry 5.0 necessitate resilient Internet of Things (IoT) networks, which are increasingly susceptible to sophisticated cyber threats. While advancements in intrusion detection systems (IDS) have improved attack detection, addressing the complexity of multi-class attack scenarios and managing minority threats remains challenging. This study proposes NFIIoT-DTL-IDS, an adaptive IoT IDS for smart network management using NetFlow and IIoT data, driven by deep transfer learning and enhanced with genetic algorithm (GA) optimization. Our framework leverages pre-trained CNN models to convert data into images, enabling effective classification. GA optimizes hyperparameters to improve model flexibility and performance, while a soft voting ensemble ensures robust aggregation of predictions. The proposed IDS achieves 100 classification accuracy among 5, 10, and 19 distinct attack classes of three IoT datasets, including two NetFlow datasets (NF-TONIoTv2 and NF-BoT-IoTv2) and one industrial-based dataset (XIIoTID), detecting threats such as DDoS, ransomware, and theft in common IoT cyberattacks, as well as MQTT subscription and crypto-ransomware attacks in industrial IoT scenarios. Additional experiments demonstrate the effectiveness of the proposed method, which exceeds the classification performance results of three baseline models, including LSTM, Transformer, and 3D CNN, by more than 37.2%, 0.25%, and 1.25%, respectively, in the maximum among the three datasets. Finally, experimental results demonstrate that NFIIoT-DTL-IDS outperforms recent state-ofthe-art solutions in terms of multiclassification accuracy. This contribution advances adaptive management frameworks for IoT security, offering scalable and high-performance solutions for intrusion detection in modern network management.