NOUIRI, LahceneDERRECHE, Seddik2024-10-212024-10-212024MM/838http://10.10.1.6:4000/handle/123456789/5657The processing of medical images, especially those obtained by magnetic resonance imaging (MRI), is important for diagnosing certain diseases such as brain tumors. In this study, we propose a method to automatically detect these tumors using the U-Net network. Our approach is based on using this architecture to extract distinctive features from MRI images, which are then used for tumor detection. The results show that our U-Net-based model achieves a detection accuracy of 95%, demonstrating its effectiveness in this context. These results show the promising potential of using U-Net to improve the early and accurate detection of brain tumors from MRI images.frMRI, U-net, segmentation, Deep learning, Brain tumorsThesis