🔬 White Blood Cell Classification using CNN

A deep learning project implementing a Convolutional Neural Network (CNN) for automated classification of white blood cells from microscopic images, achieving 95% accuracy across five cell types.

Author: Muhammad Husnain Ali

🛠️ Technologies Used

Core Technologies

• Python - Primary programming language
• TensorFlow - Deep learning framework
• Keras - Neural network API
• Google Colab - Development environment

Data Processing & Visualization

• NumPy - Numerical computing
• Pandas - Data manipulation
• PIL (Pillow) - Image processing
• Matplotlib - Data visualization
• Seaborn - Statistical visualization
• Scikit-learn - Model evaluation
• KaggleHub - Dataset management

🚀 Features

• Advanced Image Processing

  • Automatic image resizing (128x128)
  • Pixel normalization
  • Data augmentation with rotations and flips
  • Balanced class distribution

• Robust CNN Architecture

  • 4-stage convolutional network
  • Batch normalization
  • Dropout regularization
  • MaxPooling for feature selection

• Performance Optimization

  • Early stopping
  • Learning rate scheduling
  • Model checkpointing
  • GPU acceleration

• Comprehensive Evaluation

  • Confusion matrix analysis
  • ROC curve generation
  • Per-class accuracy metrics
  • Real-time prediction visualization

🏗️ Model Architecture

Input (128x128x3)
│
├── Conv Block 1 (64 filters) → BatchNorm → MaxPool → Dropout(0.25)
│
├── Conv Block 2 (128 filters) → BatchNorm → MaxPool → Dropout(0.25)
│
├── Conv Block 3 (256 filters) → BatchNorm → MaxPool → Dropout(0.25)
│
├── Conv Block 4 (512 filters) → BatchNorm → MaxPool → Dropout(0.25)
│
├── Dense(1024) → Dropout(0.5)
│
├── Dense(512) → Dropout(0.5)
│
└── Output(5) → Softmax

📊 Dataset

• Source: White Blood Cells Dataset
• Classes: 5 WBC types
  • Neutrophils (6,231 training images)
  • Lymphocytes (2,427 training images)
  • Monocytes (561 training images)
  • Eosinophils (744 training images)
  • Basophils (212 training images)
• Volume: 14,514 images
  • Training: 10,175 total
    • Training split: 8,140 (80%)
    • Validation split: 2,035 (20%)
  • Test: 4,339 (Test-A)

📈 Performance

Metric	Score
Training Accuracy	~94%
Validation Accuracy	~95%
Test Accuracy	~95%
Inference Time	<100ms
Model Size	~50MB

🚀 Quick Start

1. Open in Colab

2. Install Dependencies

!pip install kagglehub tensorflow numpy pandas matplotlib seaborn pillow scikit-learn tqdm

3. Download Dataset

import kagglehub

# Download latest version
path = kagglehub.dataset_download("masoudnickparvar/white-blood-cells-dataset")
print("Path to dataset files:", path)

4. Run the Notebook

Simply execute all cells in code.ipynb sequentially. The notebook will:

• Load and preprocess the dataset
• Display data distribution and sample images
• Build and train the CNN model
• Evaluate performance with metrics and visualizations
• Save the trained model as wbc_classification_model.keras

📦 Requirements

tensorflow>=2.8.0
numpy>=1.19.5
pandas>=1.3.0
matplotlib>=3.4.3
seaborn>=0.11.0
pillow>=8.3.0
scikit-learn>=1.0.2
tqdm>=4.62.0
kagglehub>=0.1.0

💡 Usage Tips

Training

• Use GPU runtime in Colab for faster training (Runtime → Change runtime type → GPU)
• The notebook includes data augmentation by default
• Training typically takes 10 epochs with early stopping
• Monitor validation metrics to avoid overfitting

Making Predictions

from tensorflow.keras.models import load_model
from PIL import Image
import numpy as np

# Load the trained model
model = load_model('wbc_classification_model.keras')

# Load and preprocess image
img = Image.open('path/to/image.jpg').convert('RGB')
img = img.resize((128, 128))
img_array = np.array(img) / 255.0
img_array = np.expand_dims(img_array, axis=0)

# Predict
prediction = model.predict(img_array)
classes = ["Neutrophil", "Lymphocyte", "Monocyte", "Eosinophil", "Basophil"]
predicted_class = classes[np.argmax(prediction)]
confidence = np.max(prediction)

print(f"Predicted: {predicted_class} with {confidence:.2%} confidence")

🤝 Contributing

1. Fork the repository
2. Create feature branch (git checkout -b feature/AmazingFeature)
3. Commit changes (git commit -m 'Add AmazingFeature')
4. Push to branch (git push origin feature/AmazingFeature)
5. Open Pull Request

📞 Support

For questions and support:

• Open an Issue
• Contact: m.husnainali.work@gmail.com

🙏 Acknowledgments

• Kaggle for the comprehensive dataset
• TensorFlow team for the framework
• Google Colab for free GPU resources
• All contributors to this project

---

Made with ❤️ by Muhammad Husnain Ali