Abnormal growth and excessive division of cells in the brain are the results of brain tumors. Early diagnosis of brain tumors plays a significant role in raising the likelihood of recovery and in growing the survival rate. Hence brain tumor segmentation methods become a fundamental factor in the identification of tumors. In recent ages, several approaches have been developed to automatically segment MRI brain tumors. But still, there is a need for fully functional, reasonable, and reproducible methods for accurate segmentation. Because of its high variations and uncertainties in various factors like brain tumor sizes, irregularities, positions, and orientations, multiple difficulties arise, and the development of automated and non-invasive devices becomes important in the medical field. This special issue is open to receiving several advanced methods using deep learning architectures, which have overcome this issue and outperformed other conventional techniques. More significantly, we look forward to receiving articles that contribute to the advancement in the field.