Enhanced Brain Tumor Classification Through ResNet50 Transfer Learning

This research proposes a transfer learning-based framework for the accurate and efficient classification of brain tumors utilizing Magnetic Resonance Imaging (MRI) data. A pre-trained ResNet50 model was meticulously fine-tuned on a comprehensive dataset encompassing 7023 MRI images, meticulously annotated into four distinct tumor categories. The methodology integrates image preprocessing, segmentation, feature extraction, and classification stages to optimize model performance. Experimental results underscore the efficacy of the proposed approach, demonstrating superior classification accuracy, precision, recall, and F1-scores. The model's ability to effectively differentiate between various brain tumor types holds significant potential for enhancing clinical decision-making and patient outcomes.