An Ensemble Deep Learning Approach for Brain Tumor Detection and Classification from MRI Images

Accurate and early detection of brain tumors is critical for improving treatment outcomes and survival rates. This project presents an ensemble-based deep learning approach for brain tumor detection and classification using MRI scans. Leveraging the strengths of two pre-trained Convolutional Neural Network (CNN) architectures—VGG16 and ResNet50—the system integrates their predictions through majority voting to enhance robustness and accuracy. Advanced preprocessing techniques such as skull stripping, intensity normalization, and data augmentation were employed to prepare the dataset from the BraTS 2020 collection. The ensemble model successfully classifies tumors into glioma, meningioma, pituitary, or no tumor, achieving an accuracy of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\text{9 3. 1 6\%}$</tex>, surpassing the performance of individual models. The solution is deployed via a web-based interface, providing real-time classification, chatbot support, and a donation platform for social impact.This project shows that using ensemble deep learning models can help doctors make better diagnoses and can be improved in the future by adding different types of medical scans and explainable AI.