Salhi, Mohammed Nadjib AllahLachekhab (Kahoul), Fadhila (Promoteur)2024-12-052024-12-052024https://dspace.univ-boumerdes.dz/handle/123456789/1488865 p. : ill. ; 30 cmHealthcare MRI for brain tumor is a critical aspect of modern medicine, particularly in diagnosing and treating neurological disorders. Brain tumors pose significant health risks, and early detection is key to successful treatment outcomes. Traditional diagnostic methods often involve manual interpretation of MRI images by skilled radiologists, which can be time-consuming and subject to human error. Recent advancements in medical imaging and AI have paved the way for more efficient and accurate diagnosis of brain tumors using Deep Learning algorithms. This study proposes a Deep Learning-powered MRI-based system for automated detection and localization of brain tumors. Utilize Convolutional Neural Networks (CNNs) to analyze MRI scans and classify them into two classes: "Tumor" and "No tumor." To train and evaluate the four models, a dataset comprising of MRI images with corresponding labels indicating the presence or absence of tumors is utilized and then localization of a tumor if it exists. Evaluation metrics such as accuracy, F1-score, Precision and Confusion Matrix are employed to assess the performance of the models in distinguishing between tumor and non-tumor cases. The results demonstrate the efficacy of the proposed approach in accurately identifying brain tumors from MRI scans.enProcédés de fabrication : AutomatisationApprentissage profondTumeurs cérébrales : DétectionImagerie par résonance magnétiqueRadiologuesIntelligence artificielle en médecineCNN (réseaux neuronaux convolutionnels)Pétrochimie : InstrumentationsThesis