This research introduces an accurate brain tumor detection, segmentation, and classification system in the medical field deployed on Streamlit to answer the urgent demand for a more precise and effective diagnosis of brain tumors. The existing problems in this field include a small number of available MRI modalities, a classification system that only includes a few types of brain tumors, and the division of segmentation and classification responsibilities within the systems that are currently in place. Our strategy involves neural network transfer learning and fine-tuning approach to address these problems. We have accomplished important milestones by utilizing transfer learning on top of the YOLOv8 pre-trained model provided by the Ultralytics library that achieved a classification accuracy of over 95% and a segmentation accuracy of approximately 90%.