One of the most dangerous and deadly cancers that can affect both adults and infants is the brain tumor in the world. People with this disease have died from it more frequently. Biomedical image processing facilitates the detection and identification of brain tumors using magnetic resonance imaging (MRI). The most accurate method to detect brain and spinal cord tumors is with a contrast-enhanced MRI. Using MRI, sometimes, physicians can determine whether a tumor is cancerous or not. MRI can also be used to search for indications that cancer may have spread to another body part from its original site through spread. The capacity of classifiers for MRI brain scans to extract significant features is a fundamental step. Separating a brain tumor from healthy brain tissue is a process known as brain tumor segmentation. In routine clinical practice, this knowledge is helpful for the purpose of diagnosing and treating. Even yet, the process remains difficult because of the asymmetrical shape and unclear borders of tumors. Machine learning and deep learning algorithms are used in a number of research publications to detect brain tumors. Brain cancer prediction using these methods on MRI scans can be completed rapidly and increased accuracy aids in patient treatment. Deep Learning Supported Image Interactive Medical Image Segmentation (DL-IIMIS), which incorporates CNNs into the bounding box and scribble-based pipeline, is suggested to address these challenges. A review of some of the research papers is provided in this volume on MRI brain tumor segmentation using deep learning. Recently, there have been encouraging successes in overcoming several computer vision challenges with deep learning techniques such as image classification, object detection, and semantic segmentation.