Paper Title
ENSEMBLE LEARNING VS CONVOLUTIONAL NEURAL NETWORKS FOR MULTICLASS BRAIN TUMOR CLASSIFICATION OF MRI IMAGES

Abstract
Abstract: Accurate classification of brain tumors from magnetic resonance imaging (MRI) scans is critical for selecting appropriate treatment strategies. In this work, three convolutional neural network (CNN) architectures were developed and trained on a multiclass brain tumor MRI dataset containing four types of images: no tumor, meningioma, glioma, and pituitary tumors. The CNN models achieved test accuracy scores above 99%. Ensemble techniques using average and majority voting were applied to boost performance further and integrate the three models' predictions. The ensemble approach provided a slight but noticeable improvement, with the best model reaching 99.46% accuracy. Overall, the deep CNN models demonstrated excellent capabilities for distinguishing between the multiple tumor classes from MRI scans. The ensemble method offered a way to extract incremental benefits by combining multiple trained models. Keywords - Brain Tumors, Diagnosis, Classification, Pre Trained