Brain Tumor Detection System

A full-stack web application that detects brain tumors from MRI images using a Convolutional Neural Network (CNN). This project combines a Flask backend for model inference and a React frontend for user interaction, containerized with Docker.

Features

• Upload MRI image and receive tumor prediction.
• Displays confidence levels per class.
• Drag and drop image upload with preview.
• Error handling for invalid/non-MRI images.
• Fully responsive and styled frontend.
• Scalable architecture using Docker.

---

Tech Stack

Frontend:

• React
• Axios
• Styled Components / CSS Modules

Backend:

• Python
• Flask
• TensorFlow/Keras
• Pillow
• NumPy
• Flask-CORS

DevOps:

• Docker
• Docker Compose

---

Model Information

• Trained a CNN model to classify brain MRIs into 4 categories:
  • Glioma Tumor
  • Meningioma Tumor
  • Pituitary Tumor
  • No Tumor
• Preprocessing: Images resized to 128x128 or 150x150, normalized to [0,1] range.
• Model saved in HDF5 format (brain_tumor_model.h5).
• Label encoding stored using pickle.

---

Project Structure

Brain-Tumor-Detection-System/ │ ├── backend/ │ ├── app.py │ ├── utils.py │ ├── test_api.py │ ├── model/ │ │ ├── brain_tumor_model.h5 │ │ └── label_encoder.pkl │ └── requirements.txt │ ├── frontend/ │ ├── public/ │ ├── src/ │ │ ├── components/ │ │ ├── App.js │ │ └── App.css │ └── package.json │ ├── docker-compose.yml ├── README.md

---

Installation

1. Clone the Repository

git clone https://github.com/IthavinduU/Brain-Tumor-Detection-System.git cd Brain-Tumor-Detection-System

2. Backend Setup

cd backend python -m venv venv venv\Scripts\activate # On Windows pip install -r requirements.txt python app.py

3. Frontend Setup

cd ../frontend npm install npm start

4. Run with Docker (Recommended)

docker-compose up --build

---