Brain Tumor MRI Image Classification

This project uses deep learning (CNN and Transfer Learning with MobileNetV2) to classify brain MRI images into four categories:

• Glioma
• Meningioma
• Pituitary Tumor
• No Tumor

---

Objective

To build and deploy an AI-powered classification tool that helps in medical diagnostics by predicting the type of brain tumor from MRI scans.

---

Project Structure

---

How to Run the App

Step 1: Install Required Libraries

Make sure you have Python installed. Then open terminal/command prompt and run:

Step 2: Launch the App

or simply double-click on Run_Tumor_App.bat if you are on Windows.

---

Model Details

• CNN: A custom Convolutional Neural Network was trained from scratch.
• Transfer Learning: MobileNetV2 was used to boost accuracy with pre-trained weights.
• Metrics: Accuracy, precision, recall, and loss were used to evaluate performance.
• Libraries Used: TensorFlow, OpenCV, Matplotlib, Streamlit, Scikit-learn.

---

Sample Output

• Uploads an MRI image via the Streamlit app
• Classifies it into one of the four tumor types
• Displays prediction with confidence score

---

Author

Prerna Utage
AI/ML Intern – LabMentix
GitHub: (https://github.com/prerna0412)

---

📌 Note

This project is for educational and research purposes only. Not intended for real-time clinical use.