Deep Learning-Driven Brain Tumor Sectionalisation: Leveraging YOLO11 for MRI Analysis

The radiologist has been using magnetic resonance imaging (MRI) because of its exceptional capacity to detect even the smallest abnormalities in the brain. This study contemplates an overarching examination of the object identification framework by YOU ONLY LOOK ONCE 11 (YOLO11), using a cutting-edge deep learning approach to identify and categorize brain tumors in MRI. Its flexible convolutions and lightweight hybrid neck enable precise tumor identification in MRI images, making it ideal for real-time clinical edge deployment. The dataset of brain tumor incorporates three dissimilar classes, including meningiomas, gliomas and pituitary tumors. The numerical result of YOLO11 shows excellent performance. The accuracy of tumor localization and tumor detection is 94.9% and 95.4%, respectively. The recall scores are 0.909 and 0.926 for mask mapping and box detection, respectively. In examining medical images, a larger value of mAP@.5:.95 appears to indicate greater efficiency.