Brain Tumor Segmentation Using UNet with VGG19

Accurate segmentation of brain tumors from magnetic resonance imaging (MRI) scans is essential for the diagnosis, treatment planning, and monitoring of patients with brain disorders. This paper presents a novel approach for automated brain tumor segmentation using a hybrid architecture that integrates the VGG19 convolutional neural network as the encoder and the U-Net architecture as the decoder. The VGG19 encoder is utilized for feature extraction, taking advantage of its capability to capture hierarchical features from input images. The U-Net decoder, well-known for its effectiveness in semantic segmentation tasks, refines the segmentation masks by integrating both low-level and high-level features. By combining the strengths of VGG19 in feature extraction and U-Net in spatial localization, our proposed model achieves precise and robust segmentation of brain tumors in MRI images. Extensive experiments are conducted on benchmark datasets to evaluate the proposed approach against state-of-the-art methods. The results demonstrate superior performance in terms of segmentation accuracy and computational efficiency, underscoring the potential of the VGG19-U-Net architecture.