Segmentation of Brain MRI Images Using U-Net

Abstract

A technique called U-net was primarily created for picture segmentation. Due to the significance of the U-growing net in the field of medical imaging, it is now the preferred tool for image segmentation jobs in the medical industry. Every use of U-net design, including in CT scans, MRI, X-rays, and microscopes, illustrates the efficiency of the technology. U-net is primarily thought of as a segmentation tool, but it is occasionally used for other purposes as well. It is considered the best architecture in the field of medical image segmentation. U-net strength is still increasing with so many new improvements in the architecture; this paper evaluates the growth and improvement in U-net architecture and compares it with other architectures like SegNet and LeNet. The experimental results demonstrate that the U-net model with various modifications outperforms SegNet and LeNet based approaches using the dataset BRATS2020.