Alzheimer’s Disease (AD) is a progressive neurodegenerative disorder characterized by memory loss, behavioral changes, and impaired self-care, often preceded by Mild Cognitive Impairment (MCI). Not all MCI cases progress to AD, creating a diagnostic challenge. This study proposes a novel framework for early AD diagnosis using T1-weighted Magnetic Resonance Imaging (MRI). The approach integrates the Adaptive Moving Self-Organizing Map (AMSOM), a neural network technique for unsupervised training and tissue segmentation, with K-means clustering and Principal Component Analysis (PCA) for feature selection. AMSOM dynamically updates neuron weights to improve segmentation accuracy. Classification is performed using various algorithms, evaluated on sensitivity, accuracy, precision, and similarity metrics. Compared to existing techniques such as Fuzzy C-means (FCM) and hybrid Self-Organizing Mapping-K-means (SOM-FKM), the proposed method demonstrates statistically significant improvements in tissue segmentation and classification. It achieved a mean accuracy of 99.8%, reducing the Mean Squared Error (MSE) from 2.3 to 0.44 and improving the Discriminative Overlap Index (DOI) and Tissue Clarity (TC) values to 0.435105 and 0.282381, respectively. Implemented in MATLAB, this method provides a robust, efficient framework for early AD detection, surpassing existing approaches in precision and reliability.

The COVID-19 pandemic has made face mask detection into a big thing because it is essential in public health monitoring. Meanwhile, the growing number of things that can be connected to the internet and the increasing integration of this technology mean that edge devices are now in demand for effective real-time face mask detection models. Often, existing methods require some kind of pre-installed equipment or difficult-to-manipulate environmental conditions, and computational resource constraints essentially put an end to them. In the present study, a hybrid Flame-Sailfish Optimization (HFSO)-based deep learning framework is proposed. It combines the feature extraction capabilities of ResNet50 with the efficiency of MobileNetV2. The HFSO algorithm optimizes crucial parameters such as detection thresholds and learning rates. So that the model can take full advantage of computing capacity and still operate in real time on devices with limited resources. The model was tested on three data sets - Kaggle Face Mask Detection dataset, Public Places dataset, and Public Videos dataset - achieving up to 97.5% accuracy. It outperformed the previous leader in all cases. The results prove that this framework is reliable and easily applicable for identifying people wearing masks under different conditions. However, where there is great occlusion of the face or video feed quality is bad, the model's performance will drop somewhat. Future work should focus on increasing difficulty in detections, broadening the application of this method to other health monitoring systems based on the Internet of Things, and ensuring that its robustness remains unaltered.

This research study aims to develop and implement a sustainable and effective power solution for metropolises with a power demand ranging from 2.5 to 25 MW. The primary objective is to create a hybrid energy system (HES) that integrates various power sources, such as fuel cells and solar photovoltaic (PV), with the existing utility grid, thereby satisfying energy needs while minimizing dependency on conventional fuel-based energy sources like coal and oil. To achieve this, a thorough examination of the energy demand, availability of renewable resources, and current power infrastructure is conducted. This examination focuses on optimizing the design of the HES by considering critical factors such as grid integration, power generation capacity, energy storage capacity, and control strategies. The feasibility and performance of the proposed HES are assessed using a combination of simulation tools, mathematical modeling, and system analysis methodologies. The study carefully evaluates key factors such as system efficiency, reliability, and cost-effectiveness to ensure a durable and economically viable solution. The abstract of this study emphasizes the main quantitative findings, such as a 25% decrease in energy costs and a 30% boost in overall system efficiency, positioning the HES as an attractive choice for sustainable energy management. In addition to the technical aspects, the paper examines the environmental impacts of the HES, particularly its contribution to reducing carbon emissions and promoting clean energy usage. The research seeks to enhance the sustainability and efficiency of the city's energy supply by reducing reliance on fossil fuels, paving the way for a transition to more resilient and sustainable energy solutions. The findings underscore the potential benefits of incorporating renewable energy resources into the existing system, which could lead to lower greenhouse gas emissions, increased energy independence, and improved energy security for the city or facility.

Alzheimer’s disease (AD) ranks as the sixth leading cause of death, emphasizing the need for early-stage prediction to prevent its progression. Due to the complexity and heterogeneity of medical tests, manually comparing, visualizing, and analyzing data is often difficult and time-consuming. As a result, a computational approach for accurately predicting brain changes through the classification of magnetic resonance imaging (MRI) scans becomes highly valuable, though challenging. This paper introduces a novel method for diagnosing the early stages of AD by utilizing an efficient mapping technique to differentiate between affected and normal MRI scans. The approach combines a hybrid unsupervised learning framework, specifically the adaptive moving self-organizing map (AMSOM) method integrated with Fuzzy K-means. To ensure optimal feature extraction, we introduce a hybrid learning framework that embeds feature vectors in a subspace. The analysis compares various mapping approaches to identify features linked to Alzheimer’s disease. The proposed method achieves a classification accuracy of 95.75% on the Open Access Series of Imaging Studies (OASIS) MRI brain image database, outperforming existing methods.

Breast cancer is an increasingly serious problem in contemporary society, with millions of women and men worldwide affected by the disease. While traditional cancer detection strategies are at times effective, they typically require costly and time-intensive methods for implementation. The major drawback of using conventional methods for identifying breast cancer using the available data sets is that a single algorithm is not sufficient for accurate breast cancer diagnosis due to the heterogeneity of tumors, diverse data types, pattern complexity, feature engineering and dataset overfitting. The aim is to surpass the constraints of the conventional models and develop a hybrid model. The idea is to attain higher accuracy and lower computational time than existing models. This paper presents a novel approach that combines a one-dimensional convolutional neural network (1D CNN) with long short-term memory (LSTM) for breast cancer diagnosis as the model leverages the strengths of both approaches in extracting sequential features from local data and modelling temporal dependencies and long-range relationships. To detect and classify breast cancer, the 1D CNN and LSTM are used to automatically extract and analyze features from distinguishing features from a real dataset generated from mammography reports. The proposed model is evaluated on the extracted feature of the primary available dataset consisting of mammograms from over 760 patients. The developed model achieves 99% accuracy on the test data, demonstrating its potential to provide an automated approach to breast cancer detection. The work emphasizes a significant improvement in feature extraction, accuracy, and robustness. Additionally, the proposed model’s versatility allows it to handle diverse data types, achieve better generalization and lower computational time. The model offers a high level of interpretability, which is crucial for medical professionals to understand and trust the decisionmaking process of the system. The developed hybrid model outperforms various other state-of-the-art techniques like ANN, CNN, CNN-Bi-LSTM-GRU-AM (Convolutional Neural Network-Bidirectional Long Short-Term Memory-Gated Recurrent Unit-Attention Mechanism), and CNN-GRU (Convolutional Neural Network- Gated Recurrent Unit) in terms of accuracy, feature extraction and computational time. This work emphasises the potential of 1D CNN augmented with LSTM to create an automated system for identifying breast cancer. Hence, provides a promising foundation for further development and practical usage of deep learning for automated cancer diagnosis.

Software-Defined Networking (SDN) offers a centralized network management approach that can effectively address the complex and varied traffic demands characteristic of Industrial Internet of Things (IIoT) environments by decoupling the control plane from the data plane. The centralized control architecture of SDN necessitates the performance optimization of controllers to manage diverse traffic efficiently within IIoT applications. This paper explores the criteria for selecting controllers in SDN-enabled IIoT (SD-IIoT) environments, utilizing the Less Complex Analytic Network Process (LC-ANP) to establish their prioritization. A ranking system for SD-IIoT controllers is formulated using LC-ANP, and experimental validation of this method underscores its effectiveness in optimizing controller performance. The proposed approach enhances the overall efficiency of SDN-enabled IIoT networks, as evidenced by experimental evaluations measuring delay, throughput, packet loss ratio, and jitter across five different topologies with varying nodes and edges. The experiments indicate an overall increase in the average throughput, and a decrease in delay, jitter, and packet loss ratio. The results also show that the suggested strategy and proposed controller surpass the benchmark controller in complex network topologies. These results confirm the method’s capacity to significantly improve network performance in SD-IIoT applications.

Secure and reliable exchange of information between devices is crucial for any network in the current digital world. This information is maintained on storage devices, routing devices, and communication over the cloud. Cryptographic techniques are used to ensure the secure transmission of data, ensuring the user’s privacy by storing and transmitting data in a particular format. Using encryption, only the intended user possessing the key can access the information. During data or essential transmission, the channel should be secured by using robust encryption techniques. Homomorphic Encryption (HE) techniques have been used in the past for this purpose. However, one of the flaws of the conventional HE is seen either in its slow transmission or fast key decryption. Thus, this paper proposes an optimized Homomorphic Encryption Chinese Remainder Theorem with a Rivest-Shamir-Adleman (HE-CRT-RSA) algorithm to overcome this challenge. The proposed Technique, HE-CRT-RSA, utilizes multiple keys for efficient communication and security. In addition, the performance of the HE-CRT-RSA algorithm was evaluated in comparison with the classical RSA algorithm. The result of the proposed algorithm shows performance improvement with reduced decryption time. It is observed that the proposed HE-CRT-RSA is 3–4% faster than the classical Rivest-Shamir-Adleman (RSA). The result also suggests that HE-CRT-RSA effectively enhances security issues of the cloud and helps to decrease the involvement of intruders or any third party during communication or inside the data/server centers.

Load balancing is a serious problem in cloud computing that makes it challenging to ensure the proper functioning of services contiguous to the Quality of Service, performance assessment, and compliance to the service contract as demanded from cloud service providers (CSP) to organizations. The primary objective of load balancing is to map workloads to use computing resources that significantly improve performance. Load balancing in cloud computing falls under the class of concerns defined as "NP-hard" issues due to vast solution space. Therefore it requires more time to predict the best possible solution. Few techniques can perhaps generate an ideal solution under a polynomial period to fix these issues. In previous research, Metaheuristic based strategies have been confirmed to accomplish accurate solutions under a decent period for those kinds of issues. This paper provides a comparative analysis of various metaheuristic load balancing algorithms for cloud computing based on performance factors i.e., Makespan time, degree of imbalance, response time, data center processing time, flow time, and resource utilization. The simulation results show the performance of various Meta-heuristic Load balancing methods, based on performance factors. The Particle swarm optimization method performs better in improving makespan, flow time, throughput time, response time, and degree of imbalance.

Cloud Mobile Ad-hoc Networks (Cloud-MANETs) is a framework that can access and deliver cloud services to MANET users through their smart devices. MANETs is a pool of self-organized mobile gadgets that can communicate with each other with no support from a central authority or infrastructure. The main advantage of MANETs is its ability to manage mobility while data communication between different users in the system occurs. In MANETs, clustering is an active technique used to manage mobile nodes. The security of MANETs is a key aspect for the fundamental functionality of the network. Addressing the security-related problems ensures that the confidentiality and integrity of the data transmission is secure. MANETs are highly prone to attacks because of their properties.In clustering schemes, the network is broken down to sub-networks called clusters. These clusters can have overlapping nodes or be disjointed. An enhanced node referred to asthe Cluster Head (CH) is chosen from each set to overseetasks related to routing. It decreases the member nodes’ overhead and improvesthe performance of the system. The relationship between the nodes and CH may vary randomly, leading to re-associations and re-clustering in a MANET that is clustered. An efficient and effective routing protocol is required to allow networking and to find the most suitable paths between the nodes. The networking must be spontaneous, infrastructure-less, and provide end-to-end interactions. The aim of routing is the provision of maximum network load distribution and robust networks. This study focused on the creation of a maximal route between a pair of nodes, and to ensure the appropriate and accurate delivery of the packet. The proposed solution ensured that routing can be carried out with the lowest bandwidth consumption. Compared to existing protocols, the proposed solution had a control overhead of 24, packet delivery ratio of 81, the lowest average end-to-end delay of 6, and an improved throughput of 80,000, thereby enhancing the output of the network. Our result shows that multipath routing enables the network to identify alternate paths connecting the destination and source. Routing is required to conserve energy and for optimum bandwidth utilization.

Cloud computing is an extremely important infrastructure used to perform tasks over processing units. Despite its numerous benefits, a cloud platform has several challenges preventing it from carrying out an efficient workflow submission. One of these is linked to task scheduling. An optimization problem related to this is the maximal determination of cloud computing scheduling criteria. Existing methods have been unable to find the quality of service (QoS) limits of users- like meeting the economic restrictions and reduction of the makespan. Of all these methods, the Heterogeneous Earliest Finish Time (HEFT) algorithm produces the maximum outcomes for scheduling tasks in a heterogeneous environment in a reduced time. Reviewed literature proves that HEFT is efficient in terms of execution time and quality of schedule. The HEFT algorithm makes use of average communication and computation costs as weights in the DAG. In some cases, however, the average cost of computation and selecting the first empty slot may not be enough for a good solution to be produced. In this paper, we propose different HEFT algorithm versions altered to produce improved results. In the first stage (rank generation), we execute several methodologies to calculate the ranks, and in the second stage, we alter how the empty slots are selected for the task scheduling. These alterations do not add any cost to the primary HEFT algorithm, and reduce the makespan of the virtual machines’ workflow submissions. Our findings suggest that the altered versions of the HEFT algorithm have a better performance than the basic HEFT algorithm regarding decreased schedule length of the workflow problems.