Brain Tumor Classification

This project focuses on classifying brain tumors using deep learning models. Medical image data in .mat format was preprocessed, resized, and converted to PNG format for training. The dataset was structured into classes and processed using TensorFlow.

A Convolutional Neural Network (CNN) was built to classify tumor types. To enhance transparency and model explainability, Grad-CAM (Gradient-weighted Class Activation Mapping) was applied to visualize which regions of the brain influenced the model’s decisions.

Key Features:

1) Data preparation from .mat files

2) Image resizing and formatting (240×240)

3) CNN model for tumor classification

4) Grad-CAM visualizations for interpretability

5) TensorFlow and Keras for model development

This project demonstrates practical experience in deep learning, medical imaging, and interpretable AI, and can be adapted to other medical diagnosis tasks.