Description

MRI Brain Tumor Detection is a deep learning project that uses a Convolutional Neural Network (CNN) to classify brain MRI scans into tumor categories such as glioma, meningioma, pituitary, or no tumor. The model is trained and evaluated in a Jupyter Notebook and deployed through a Flask web application for real-time predictions, demonstrating an end-to-end medical imaging and deployment workflow.

Key Features

• CNN-Based Tumor Classification for MRI brain images
• End-to-End ML Pipeline from preprocessing and training to deployment
• Flask Web Application for real-time model predictions
• Multi-Class Detection (glioma, meningioma, pituitary, no tumor)
• Model Evaluation & Visualization with accuracy, loss, and confusion matrix
• Deployment-Ready Setup tested on Render.com
• Logging & Error Handling for debugging and reliability

Technologies Used

• Python 3.11 – Core development language
• TensorFlow & Keras – CNN model training and inference
• Flask – Web application framework
• PIL (Pillow) – Image loading and processing
• NumPy & Pandas – Data handling and preprocessing
• Matplotlib & Seaborn – Model evaluation and visualization
• Jupyter Notebook – Experimentation and model development

Design Highlights

• Notebook-to-Production Workflow bridging research and real-world use
• Deep CNN Architecture optimized for medical image classification
• Separation of Concerns between model training and web inference
• Resource-Aware Deployment designed for cloud free-tier constraints
• Scalable Design ready for future user image uploads
• Clear Evaluation Pipeline with interpretable metrics and visuals