🧠 Brain Tumor Classification Using CNNs

This repository contains multiple deep learning models to classify brain tumors into Meningioma, Glioma, and Pituitary tumor types using MRI images. The best-performing model is a custom-built CNN achieving an impressive 95.11% accuracy.

---

📂 Dataset

• Classes (after One-Hot Encoding):

  Label	Class	Count
  0	Meningioma	708
  1	Glioma	1426
  2	Pituitary Tumor	930

---

🏗️ Models Compared

Model Name	Description
MODEL1: MobileNetV2	Lightweight CNN, pretrained on ImageNet, good for small datasets and fast inference
MODEL2: ResNet50	Deep residual network, ImageNet pretrained, better for deeper feature learning
MODEL3: EfficientNetB0	Balanced in speed and accuracy, state-of-the-art architecture
MODEL4: Custom CNN ✅	Built from scratch, fast experimentation, fine-tuned layers – Best Performance
MODEL5: Custom CNN + BN	Same as above but with added BatchNormalization layers for better generalization

---

✅ Best Performing Model: Custom CNN

• Built from scratch using Keras
• Fine-tuned with additional optimizations
• Accuracy: 95.11%
• Output layer: Dense(3, activation='softmax')

---

📊 Results

Model	Accuracy
MobileNetV2	~88%
ResNet50	~91%
EfficientNetB0	~92.5%
Custom CNN (Model 4)	95.11%
Custom CNN + BN	~93%

---

🛠️ Installation

# Clone the repository
git clone https://github.com/fazeelibtesam/brain-tumor-classification-cnn.git
cd brain-tumor-classification-cnn

# Create and activate a virtual environment (optional but recommended)
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

---

🚀 How to Run

1. Prepare Dataset Ensure the dataset is structured as:

   dataset/
   ├── glioma/
   ├── meningioma/
   └── pituitary/
   

2. Train a Model Modify and run the desired script:

   python train_custom_cnn.py  # For Model 4

3. Evaluate the Model

   python evaluate_model.py

---

📈 Future Work

• Implement advanced data augmentation
• Test with more EfficientNet variants (B1–B3)

---

📄 License

This project is licensed under the Apache2.0 License.

---

🙌 Acknowledgments

• MRI Dataset from Cheng, Jun (2017)
• Pretrained models via keras.applications

---

🤝 Contact for Collaboration

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.

I'm open to feedback, collaboration, or ideas for improvements. Feel free to reach out!

• 📧 Email: ibtesamfazeel@gmail.com
• 💼 LinkedIn: https://linkedin.com/in/fazeel-ibtesam
• 🐙 GitHub: https://github.com/fazeelibtesam

Feel free to fork, star ⭐, and follow for updates!

---