Deep Learning to Revolutionize Brain Tumor Detection and Classification Using Multi-Plane MRI Scans

Early and accurate detection of brain tumors is essential for guiding effective treatment, and MRI imaging has become a crucial tool due to its detailed and high-resolution visuals. However, detecting and classifying brain tumors in MRIs can be difficult, as the images differ according to the axial, coronal, and sagittal planes and may contain noise that complicates identification. In this study, we explore the powerful YOLOv5 model to identify brain tumors through these three MRI views. By categorizing both positive and negative tumor cases into six distinct classes and integrating background images, the model's accuracy and adaptability have been enhanced. The YOLOv5 small weighted model has delivered exceptional results, achieving a mean average precision (mAP50) of 0.96, a recall of 0.90, and an overall precision of 0.91 on unseen test data. These metrics underscore YOLOv5's remarkable accuracy in tumor localization across various MRI planes, highlighting its potential to revolutionize clinical diagnostics and treatment planning. Along with better treatment for patients, this model provides doctors with a powerful tool for accurate case analysis.