The initial testing and treatment planning of brain tumors using MRI scans depend on brain tumor segmentation, which has a direct effect on patient outcomes. Because manual segmentation takes a lot of time and arduous, automated techniques are required for efficiency. Our paper focuses on reviewing MRI-based brain tumor segmentation techniques, particularly emphasizing recent advancements in deep learning These deep learning techniques are becoming more and more well-liked because they can process complex image data efficiently and produce cutting-edge outcomes. We begin by introducing the significance of brain tumor segmentation and conventional methodologies before delving into a detailed discussion of cutting-edge deep learning algorithms. By critically evaluating these methods, we aim to elucidate their strengths and limitations, paving the way for their integration into routine clinical practice. Through our comprehensive review, we strive to contribute to the advancement of brain tumor diagnosis, ultimately enhancing patient care in neuro-oncology.