Eritrea Institute of Technology

Fungal dynamicity drives rapid adaptation to environmental changes, contributing to the development of antifungal resistance, phenotypic switching, and the emergence of opportunistic pathogenicity. Genomic plasticity or flexibility is a hallmark of fungal evolution, driving fungal diversity through the formation of diverse ecological niches and relationships such as mutualistic, parasitic, and opportunistic, thereby enabling fungi to face external pressures. This review proposes a conceptual framework that integrates three interlinked dimensions of fungal adaptability: genomic plasticity, phenotypic adaptability, and ecological resilience. It explores the mechanisms underlying genotypic and phenotypic changes and the implications of these changes on fungal virulence, environmental adaptation, drug resistance, and biotechnological applications. In addition, the link between environmental pressures and genomic changes is highlighted. A deeper understanding of these genomic dynamics contributes to the design of effective drug therapies so as to reduce the rising antifungal drug resistance issue in the health and agricultural sectors.

Intelligent scheduling of heterogeneous cloud computing systems is the key to energy efficiency, optimal performance and cost and environmental impact reduction. In this paper, a novel ant colony inspired cooperative foraging behavior based meta-heuristic optimization framework named EcoTaskOpt is proposed for the efficient task scheduling among the heterogeneous cloud resources. EcoTaskOpt combines Ant Colony Optimization (ACO) and Particle Swarm Optimization (PSO) algorithm in a hybrid model, where the updates of pheromone information and particle velocity are dynamically adjusted to achieve the optimization of task to resource mappings with the balance between exploration and exploitation in complex scheduling scenarios. For evaluation, a benchmark data set is created using CloudSim, including real-time Google Cluster traces, synthetic task graphs with different levels of computational complexity, server power profile and thermal dynamics related to workload intensity. Simulation results show that compared with the conventional heuristics, deep reinforcement learning and hybrid evolutionary algorithms, EcoTaskOpt reduces energy consumption by up to 42

Melanoma remains one of the most aggressive forms of skin cancer, and early diagnosis is critical to improving patient survival. This study presents an Adaptive Hybrid AI Framework (AHA-Net) designed for accurate and interpretable skin lesion segmentation and melanoma classification. The proposed architecture enhances a modified UNet + + backbone with an Adaptive Scaled Dot Attention Mechanism (A-SDAM) that dynamically regulates attention sharpness across multiple lesion scales. A Residual Cross-Attention Bridge (RCAB) enables effective contextual fusion between segmentation and classification pathways, while Vision Transformer (ViT)–based multi-scale attention layers capture both local and global dependencies. A hybrid CNN–ViT classifier, refined with dynamically weighted SVM post-classification, improves decision boundary precision under class imbalance. Furthermore, self-distillation between ViT layers enhances feature coherence and cross-dataset generalization. The framework was rigorously evaluated across four benchmark datasets i.e., HAM10000, ISIC 2019; ISIC 2020, and PH2, representing diverse lesion types and imaging conditions. Experimental results demonstrate that Proposed AHA-Net consistently outperforms existing state-of-the-art architectures, including U-Net, ResNet, UNet++, and TransUNet, in both segmentation and classification tasks. Statistical analysis confirms the significance and reproducibility of performance gains (p < 0.05). Quantitative explainability assessment using Grad-CAM shows that the model’s attention maps align closely with clinically relevant melanoma features such as irregular borders, color heterogeneity, and asymmetric structures. These results establish Proposed AHA-Net as a robust, generalizable, and explainable AI framework with strong potential for integration into real-world dermatological diagnostic workflows.

The rapid expansion of consumer electronics has created an urgent need for advanced solutions to critical challenges such as predictive maintenance, user personalization, and device security. Traditional models often struggle to address these issues due to their limited adaptability and performance. This paper introduces GenAI-A, an innovative AI model that integrates Generative Adversarial Networks (GANs), Variational Autoencoders (VAEs), Dynamic Recommendation Algorithms (DRA), and anomaly detection techniques to provide a comprehensive solution for consumer electronics. By leveraging the strengths of generative AI and self-renewal capabilities, GenAI-A enhances predictive maintenance, optimizes user experience, and strengthens biometric security. The model’s hybrid architecture utilizes GANs for realistic data representation, VAEs for the generation of complex data distributions, and DRA for real-time user personalization. The novelty of GenAI-A lies in its cross-regularized coupling between the GAN and VAE modules, where latent features are jointly optimized through a shared loss function to achieve consistent generative–representational learning. Unlike conventional hybrids that treat these models independently, GenAI-A introduces a dynamic feedback mechanism in which the DRA and anomaly detection modules operate directly in the shared latent space, enabling self-adaptive personalization and continual refinement of generative outputs. Experimental validation across four real-world datasets, i.e., Smartphone Sensor, Labelled Faces in the Wild (LFW), Pecan Street Energy Consumption, and SECOM Manufacturing, demonstrates significant improvements in device uptime, user engagement, and biometric security, with a notable reduction in false positives. Unlike existing hybrid generative models, GenAI-A introduces a novel integration of these components in a dynamic, self-learning system that adapts in real time to evolving user behaviors and device conditions. This unique combination of techniques sets GenAI-A apart from traditional approaches, establishing a new benchmark for AI-driven solutions in consumer electronics.

Recent advances have been made towards this goal, owing to the explosion of resource-constrained IoT devices: wearables, smart sensors and embedded systems, raising the need for efficient cryptographic primitives that offer high security margins while dramatically reducing their resource needs. Existing algorithms such as DES, 3DES, AES and Blowfish and RSA are secure but too resource-hungry for these devices in terms of computation and memory. In order to overcome these drawbacks, we present a Modified Chaos-Driven Enhanced Cryptographic Framework that is intentionally devoted to the security of IoT in resource-limited settings. It is a way to improve the classic logistic map by sharing entropy from dynamic key generation using smaller seeds, which makes the load faster for computation and stronger key unpredictability. The chaotic nature of the sensitive dependence on initial conditions means that the keys are highly unpredictable, which enhances confusion and diffusion properties of secure encryption. The model uses XOR-based encryption and modular arithmetic for nonlinear transformations in order to enhance security with reduced computational complexity. The framework is efficient in resource-constrained environments that are based on a lightweight symmetric encryption approach. Permuting the indices additionally increases diffusion, thereby increasing the system’s resistance to cryptanalysis. With the threat posed by quantum computing in mind, we incorporate Post-Quantum Cryptography (PQC) schemes like lattice-based cryptography and hash-based digital signatures into the framework to future-proof it. This hybrid scheme provides security against classical and quantum adversaries. Additionally, the lightweight nature of the framework allows it to run efficiently on low-power devices with minimal memory requirements, making it a feasible solution for wide deployment in real-time IoT applications such as smart cities, healthcare, and industrial IoT systems. This ensures scalability and adaptability to the growing and diverse needs of the IoT landscape.