• Likhith Raju P
• Mr. Umesh M
• Keerthi N
• Pavan Kumar
• Kushal Kumar K M

This project focuses on creating an MRI-based automated system to classify medulloblastoma and distinguish it from other brain conditions, including glioma, pituitary tumors, meningioma, and non-tumor cases. By applying deep learning methods, particularly Convolutional Neural Networks (CNNs), the system analyzes MRI images to extract key features that help in identifying each condition. Data augmentation and transfer learning, using models like ResNet, enhance classification accuracy and reduce training time. Explainability tools such as Grad-CAM add transparency to the model’s decisions, supporting its clinical credibility. This tool is designed to support radiologists in making quicker and additional accurate diagnoses, addressing a few of the challenges of manual interpretation. Future advancements will consider integrating multi-modal data and applying 3D CNNs to further improve classification accuracy and expand its use in neuro-oncology.

Deep Learning, Convolutional Neural Network(CNN), Image Processing, Automated Diagnosis, Tumor Detection, Predictive Analytics, Medical Imaging

"A Survey on MRI-based Medullablastoma Classification", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 11, page no.f341-f345, November-2024, Available :http://www.jetir.org/papers/JETIR2411538.pdf

"A Survey on MRI-based Medullablastoma Classification", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 11, page no. ppf341-f345, November-2024, Available at : http://www.jetir.org/papers/JETIR2411538.pdf