🩺 Chronic Kidney Disease Prediction using Neural Networks

🎯 Overview

This project leverages the power of Neural Networks to predict Chronic Kidney Disease (CKD) based on a comprehensive set of health records. Utilizing the UCI Chronic Kidney Disease dataset, this solution provides a robust classification model, complete with detailed data preprocessing, exploratory data analysis, and model interpretability.

🔬 Medical AI at its finest - Helping healthcare professionals make informed decisions through machine learning

✨ Key Features

🔧 Data Processing & Engineering

• Smart Missing Value Handling: Robust imputation using mean/median for numerical and mode for categorical data
• Data Cleaning: Automated detection and correction of inconsistent data entries
• Feature Engineering: Label encoding, Min-Max scaling, and PCA dimensionality reduction
• Class Balance: RandomOverSampler to handle imbalanced datasets

📊 Comprehensive Analysis

• Exploratory Data Analysis: Interactive visualizations including pairplots, distribution plots, and boxplots
• Outlier Detection: Advanced statistical methods to identify and handle outliers
• Correlation Analysis: Heatmaps to understand feature relationships

🧠 Advanced Neural Network

• Sequential Keras Model: Optimized architecture with Dense layers and Dropout
• Hyperparameter Tuning: Fine-tuned for optimal performance
• Early Stopping: Prevents overfitting during training

📈 Model Evaluation & Interpretability

• Multi-metric Evaluation: Accuracy, F1-Score, AUC, Precision-Recall curves
• Confusion Matrix: Detailed classification performance analysis
• SHAP Integration: Explainable AI for transparent decision-making
• Feature Importance: Understanding which health indicators matter most

📊 Dataset Information

Attribute	Description	Type
Dataset Source	UCI ML Repository	-
Instances	400 patients	-
Features	24 health indicators	Mixed
Target	CKD vs Non-CKD	Binary

🏥 Health Indicators Include:

• Age, Blood Pressure, Specific Gravity
• Albumin, Sugar, Red Blood Cells
• Hemoglobin, Packed Cell Volume
• White Blood Cell Count, and more...

🚀 Quick Start

Prerequisites

Python 3.7+
pip (Python package manager)

Installation

# Clone the repository
git clone https://github.com/ryuzaki-ved/ckd_detection
cd chronic-kidney-disease-prediction

# Install required packages
pip install pandas numpy matplotlib seaborn scikit-learn imbalanced-learn tensorflow shap opencv-python

🏃‍♀️ Running the Model

# Run the main prediction script
python code1.py

# For step-by-step analysis
python converted_code.py

📁 Project Structure

📦 chronic-kidney-disease-prediction/
├── 📄 README.md                    # You are here!
├── 📄 LICENSE                      # MIT License
├── 🐍 code1.py                     # Main neural network implementation
├── 🐍 converted_code.py            # Step-by-step analysis
├── 🐍 digital_tissue_mapping.py    # Additional tissue analysis
├── 📊 kidney_disease.csv           # Original dataset
├── 📊 Final_pre_processing_data.csv # Processed dataset
├── 🤖 kidney_disease_model.h5      # Trained model
├── 📝 step_by_step.txt             # Detailed methodology
├── 🔗 helpful_links.txt            # Additional resources
└── 📁 output_images/               # Generated visualizations

📈 Model Performance

Our Neural Network achieves impressive results:

• 🎯 High Accuracy: Optimized for medical diagnosis standards
• ⚖️ Balanced Performance: Handles both CKD and Non-CKD cases effectively
• 🔍 Interpretable: SHAP values explain each prediction
• 📊 Comprehensive Metrics: F1-Score, Precision, Recall, AUC

Run the model to see exact performance metrics!

🔬 Advanced Features

🧪 SHAP Explainability

# Generate SHAP explanations
explainer = shap.KernelExplainer(model.predict, x_train)
shap_values = explainer.shap_values(x_test[:100])
shap.summary_plot(shap_values, x_test[:100])

📊 Visualization Gallery

The project generates various visualizations:

• 📈 Training history plots
• 🎯 ROC curves
• 📊 Precision-Recall curves
• 🔥 Confusion matrices
• 🌟 SHAP feature importance plots

🛠️ Usage Examples

Basic Prediction

# Load and preprocess data
df = pd.read_csv("kidney_disease.csv")
# ... preprocessing steps ...

# Train model
model = create_neural_network()
history = model.fit(x_train, y_train, epochs=20)

# Make predictions
predictions = model.predict(x_test)

Model Interpretation

# Explain individual predictions
explain_prediction(model, explainer, sample_index=5)

🤝 Contributing

We welcome contributions! Here's how you can help:

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

💡 Ideas for Contributions

• Additional ML algorithms comparison
• Web interface for predictions
• Mobile app integration
• Enhanced visualizations
• Performance optimizations

📚 Documentation

• 📖 Step-by-Step Guide: See step_by_step.txt for detailed methodology
• 🔗 Helpful Resources: Check helpful_links.txt for additional materials
• 📊 Research Papers: Explore the research_papers/ directory

🏆 Acknowledgements

• 🎓 UCI Machine Learning Repository for the comprehensive dataset
• 🧠 TensorFlow Team for the amazing deep learning framework
• 📊 SHAP Contributors for explainable AI capabilities
• 👥 Open Source Community for continuous inspiration

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

📞 Support & Contact

• 🐛 Issues: GitHub Issues
• 💬 Discussions: GitHub Discussions
• 📧 Email: [email protected]

🌟 Star History

If this project helped you, please consider giving it a ⭐!

---

Made with ❤️ for better healthcare through AI

⬆ Back to Top