Brain Tumor MRI
Description
Brain Tumor MRI Classification Dataset — 4-Class Medical Imaging Dataset 1. Overview A curated medical imaging dataset containing 7,023 MRI scans for automatic brain tumor classification. The dataset supports 4 classes: Glioma, Meningioma, Pituitary, and No Tumor. It is designed for deep learning-based medical image analysis and is used to train a high-accuracy CNN + Vision Transformer + Multi-Agent classification pipeline. 2. Dataset Statistics Total Images: 7,023 Training: 5,712 images (81.3%) Testing: 1,311 images (18.7%) 3. Class Distribution Training Set: Glioma 1,321 (23.1%), Meningioma 1,339 (23.4%), No Tumor 1,595 (27.9%), Pituitary 1,457 (25.5%) Testing Set: Glioma 300 (22.9%), Meningioma 306 (23.3%), No Tumor 405 (30.9%), Pituitary 300 (22.9%) Overall: Glioma 1,621 (23.1%), Meningioma 1,645 (23.4%), No Tumor 2,000 (28.5%), Pituitary 1,757 (25.0%)
Files
Steps to reproduce
This directory contains the Brain Tumor MRI Classification Dataset, a curated collection of 7,023 MRI brain scan images organized into training and testing splits. The dataset is structured into four diagnostic classes: Glioma Meningioma Pituitary Tumor No Tumor All images are stored in class-specific subfolders under the Training and Testing directories. The dataset is used for developing and evaluating deep learning models for multi-class brain tumor classification, including CNN, Vision Transformer, and multi-agent architectures.