Brain Tumor Dataset (MRI Scans)

The Brain MRI dataset is a meticulously curated collection of 7,023 brain MRI images, designed to aid in developing and training advanced brain tumor detection models. This dataset provides a balanced distribution of images, enabling precise analysis and model performance evaluation.

Dataset Overview

1. Dataset Size and Split

   • Total Images: 7,023 brain MRI scans.
   • Split: 80:20 ratio for training and evaluation sets.
     • Training Set: 5,618 images.
     • Evaluation Set: 1,405 images.

2. Image Categories
   Each subset includes images categorized into two classes:

   • With Tumor: Includes MRI scans showing tumors.
   • Without Tumor: Includes MRI scans of healthy brains.

3. Tumor Types Covered
   The dataset features MRI scans of brains affected by the following tumor types:

   • Glioma: A type of tumor that occurs in the brain or spinal cord.
   • Meningioma: Tumors that arise from the meninges, the membranes covering the brain and spinal cord.
   • Pituitary Tumors: Abnormal growths in the pituitary gland.

4. Healthy Brain Scans
   The dataset also includes MRI images of healthy individuals, ensuring comprehensive representation for training models.

Applications of the Brain MRI Dataset

• Brain Tumor Detection: Train AI models to identify the presence or absence of tumors.
• Tumor Classification: Develop models to classify tumors into glioma, meningioma, or pituitary categories.
• Medical Research: Facilitate research in radiology and oncology for better diagnostic techniques.
• Healthcare AI Development: Improve diagnostic accuracy and support automated medical imaging systems.