Advancing Brain Tumour Detection and Classification: Knowledge Distilled ResNeXt Model for Multi-Class MRI Analysis

Accurate and timely diagnosis of brain tumors is crucial for optimal patient outcomes. Despite advancements in medical imaging and deep learning, the accurate classification of brain tumors remains a significant challenge. Existing methods, including CNNs and VGG16, often struggle to differentiate between tumor types and capture subtle radiological features. To address these limitations, we propose a novel Knowledge Distilled ResNeXt architecture. By transferring knowledge from a complex teacher model, our model effectively learns discriminative features and improves classification accuracy. Our comprehensive experiments demonstrate the superiority of the Knowledge Distilled ResNeXt in classifying brain tumors (glioma, meningioma, pituitary tumor, and no tumor) compared to state-of-the-art methods. This research contributes to the development of more effective diagnostic tools and improved patient care.