Semantic Segmentation of Brain Tumor Images Using Attention-based Residual U-Net Model

In medical image analysis, brain tumor segmentation is crucial for making an accurate diagnosis and formulating a treatment plan for patients with MRI brain tumors images.Semantic segmentation of brain tumor images plays a crucial role in medical image analysis for accurate diagnosis and treatment planning. In this paper, we introduced an attention-based residual U-Net model by combining the advantages of attention processes and residual connections for tumor segmentation and conduct a comparative analysis of the effects of activation functions, ReLU and Leaky ReLU. We evaluate the performance of the proposed model on a benchmark dataset, comparing the results obtained using both activation functions. The experimental analysis shows the impact of the activation function on the model’s performance in terms of segmentation accuracy, loss function and provides insights into the benefits of utilizing Leaky ReLU .