Precise segmentation of gliomas in magnetic resonance imaging (MRI) is essential for enhancing the diagnosis and treatment of brain cancer. This study employs advanced deep learning techniques, specifically a modified U-Net convolutional neural network, to identify and segment glioma-affected regions in T1-weighted MRI images. The model demonstrated high accuracy in segmentation tasks, outperforming existing methods in accuracy and processing time. Trained and evaluated on a comprehensive clinical dataset, our approach shows substantial promise in assisting specialists in analyzing brain tumors more effectively and optimizing radiology workflows. These findings highlight the innovative potential of deep learning in medical imaging.