Switch getting to know has become increasingly famous in many aspects of pc vision, including picture segmentation. This paper explores switch-gaining knowledge for automated brain tumor segmentation using a deep getting-to-know structure known as U-net. U-internet is a convolution neural community version that has been used efficiently to ramify scientific image segmentation obligations in the beyond. This work utilizes a U-internet model skilled on a huge-scale brain tumor segmentation dataset. The version plays nicely at the utility of brain tumor segmentation, achieving an average cube coefficient of zero. Ninety-three on check scans. Moreover, qualitative consequences display that the U-net version produces segmentation masks with few false-nice mistakes. These consequences reveal the capability of switch getting to know for automatic mind tumor segmentation.