Optimizing Brain Tumor Classification Using YOLOv8: A Comparative Study with CNN Architectures on MRI Scans

Accurate diagnosis of brain tumors is crucial for effective treatment, and advancements in artificial intelligence (AI) are transforming this process. This study adapts You Only Look Once version 8 (YOLOv8), a state-of-the-art deep learning model, for brain tumor classification using MRI scans across four categories: glioma, meningioma, pituitary tumors, and no tumor. Trained over 20 epochs with preprocessing techniques like resizing and normalization, YOLOv8 achieved a top-1 accuracy of 99.12%, outperforming traditional CNNs (94-96% accuracy). The model demonstrated faster inference, superior generalization, and consistent reductions in training and validation loss, underscoring its robustness. These results position YOLOv8 as a promising tool for early, precise, and automated tumor diagnosis, with potential for clinical deployment. Future work will focus on dataset expansion and further optimization to enhance real-world applicability