http://dspace.univ-mascara.dz:8080/jspui/handle/123456789/10 2026-05-20T04:23:33Z 2026-05-20T04:23:33Z KHALDI, Brahim http://dspace.univ-mascara.dz:8080/jspui/handle/123456789/1441 2026-05-06T13:17:54Z 2026-05-06T00:00:00Z

Titre: Design of Bio-Inspired Metaheuristics for Medical Image Classification Auteur(s): KHALDI, Brahim Résumé: Breast cancer diagnosis relies fundamentally on histopathological examination of tissue samples; however, manual microscopic evaluation by pathologists is subjective, time-consuming, and prone to inter-observer variability. While deep learning has demonstrated remarkable success in medical image analysis, conventional single-architecture models exhibit inherent limitations. Convolutional neural networks excel at local feature extraction but fail to capture global contextual relationships, whereas Vision Transformers effectively model long-range dependencies but may be inefficient in representing fine-grained local details that are critical for histopathological interpretation. Moreover, optimizing deep learning models through hyperparameter tuning and feature selection remains computationally expensive and often suboptimal when relying solely on gradient-based optimization techniques. This thesis proposes HNet, a novel hybrid deep learning architecture that strategically integrates three complementary neural paradigms for histopathological image classification. HNet combines EfficientNet for efficient local feature extraction from tissue morphology, an Advanced Vision Transformer (AVT) for capturing global contextual relationships and long-range tissue patterns, and Capsule Networks (CapsNet) for explicitly modeling spatial hierarchies and part-whole relationships inherent in tissue organization. To address the challenges of high-dimensional feature spaces and suboptimal parameter tuning, a Genetic Algorithm (GA)-based feature selection mechanism is incorporated as a critical preprocessing step between the concatenated EfficientNet-AVT feature representations and the CapsNet input. This metaheuristic-driven optimization enables automatic identification of the most discriminative features while reducing dimensionality and computational complexity. Comprehensive experimental validation is conducted on the BreakHis dataset, encompassing binary (benign vs. malignant) breast cancer classification tasks. Detailed ablation studies quantify the individual contributions of each architectural component as well as the impact of GA-based feature selection on overall performance. Comparative evaluation against recent state-of-the-art methods demonstrates that the proposed HNet architecture achieves superior classification accuracy, sensitivity, specificity, and F1-score, establishing a new benchmark for breast cancer histopathology classification. The integration of metaheuristic-driven optimization with hybrid deep learning significantly enhances model robustness, generalization to unseen data, and computational efficiency compared to non- optimized approaches. Beyond empirical performance gains, this work demonstrates how hybrid metaheuristic, deep learning frameworks can be effectively integrated into clinical workflows by addressing real-world deployment constraints, including inference latency, computational resource utilization, and model interpretability. The proposed methodology provides a generalizable template for applying metaheuristic-optimized 5 hybrid deep learning to a wide range of medical image classification tasks, offering substantial potential for advancing automated computer-aided diagnosis systems and digital pathology.

2026-05-06T00:00:00Z MELKAOUI, Chikh http://dspace.univ-mascara.dz:8080/jspui/handle/123456789/1439 2026-05-04T13:20:41Z 2026-05-04T00:00:00Z

Titre: Recherche des contaminants chimiques (Métaux lourds et HPA) dans des plantes médicinales et épices utilisées par la population du sud-ouest Algérien Auteur(s): MELKAOUI, Chikh Résumé: Dans cette étude, les niveaux et les évaluations des risques pour la santé de 15 éléments, à la fois essentiels et potentiellement toxiques, ont été déterminés dans 9 plantes médicinales collectées dans leur environnement naturel des régions sahariennes algériennes. De plus, 9 métaux lourds ont été analysés dans 5 épices collectées dans trois régions différentes du sud-ouest de l'Algérie (Naâma, Béchar et Adrar). Les concentrations des éléments ont été mesurées à l’aide de la spectrométrie d’émission optique à plasma induit (ICP-OES) après une digestion acide (69 % HNO₃ et 30 % H₂O₂) dans des conditions optimisées, ainsi que par spectrométrie de fluorescence des rayons X (XRF). L’analyse des tests enrichis et des matériaux de référence a montré une excellente linéarité des courbes d’étalonnage, avec des coefficients de corrélation (R²) allant de 0,9994 à 0,9999. Les pourcentages de récupération calculés pour la méthode de digestion des échantillons de plantes médicinales se situaient entre 84 % et 106 %, ce qui est conforme aux normes acceptables pour tous les métaux, indiquant l’efficacité de la méthode optimisée. Les niveaux de concentration des métaux lourds obtenus étaient inférieurs aux valeurs journalières recommandées par les directives de la FAO/OMS (2015). Cependant, toutes les valeurs du quotient de danger (HQ) calculées pour les trois sites étaient inférieures à la limite permise (1,0), indiquant l’absence de risque significatif pour la santé non cancérigène lié à la consommation des épices et des plantes médicinales des zones étudiées.

2026-05-04T00:00:00Z ZAGANE, Mohammed http://dspace.univ-mascara.dz:8080/jspui/handle/123456789/1432 2026-05-03T10:20:34Z 2026-05-03T00:00:00Z

Titre: Algorithmics and Data Structures 1 : Lecture notes and solved exercises Auteur(s): ZAGANE, Mohammed

2026-05-03T00:00:00Z Khatemi, Fouad http://dspace.univ-mascara.dz:8080/jspui/handle/123456789/1429 2026-04-27T08:50:38Z 2026-04-27T00:00:00Z

Titre: Improving IoT Network Security Using Deep Learning and Hybrid Feature Selection Techniques Auteur(s): Khatemi, Fouad Résumé: The rapid expansion of the Internet of Things (IoT) has introduced significant security challenges due to heterogeneous devices, high-dimensional data, and evolving cyberattacks. Traditional intrusion detection systems (IDSs) often struggle to achieve high accuracy while maintaining computational efficiency in such environments. This thesis proposes an enhanced deep learning-based IDS that integrates a dual statistical feature selection framework to improve detection performance and reduce computational complexity. The proposed approach combines Pearson correlation and Mutual Information in a two-stage feature selection process to identify the most relevant features from IoT network traffic. This framework is integrated with a fully connected deep neural network designed to learn complex intrusion patterns. Two IDS architectures are evaluated using the NSL-KDD and RT-IoT2022 datasets. Experimental results show that the proposed system achieves high detection accuracy, exceeding 99% in most scenarios, while reducing training time. Comparative analysis with existing machine learning and deep learning approaches confirms the effectiveness and robustness of the proposed solution for IoT intrusion detection.

2026-04-27T00:00:00Z