A comprehensive AI-powered health screening assistant that predicts the likelihood of Diabetes, Heart Disease, and Parkinson's Disease using machine learning models. Built with Streamlit for an intuitive web interface and deployed with pre-trained models for real-time predictions.
- Multi-Disease Prediction: Supports prediction for three critical health conditions
- Interactive Web Interface: User-friendly Streamlit application with intuitive navigation
- Real-time Predictions: Instant results using pre-trained machine learning models
- Sample Data Provided: Test cases included for each disease prediction
- Standardized Input Processing: Features are automatically scaled for optimal model performance
Predicts diabetes risk based on:
- Number of pregnancies
- Glucose level
- Blood pressure
- Skin thickness
- Insulin level
- BMI (Body Mass Index)
- Diabetes Pedigree Function
- Age
Assesses cardiovascular risk using:
- Age and gender
- Chest pain type
- Resting blood pressure
- Cholesterol levels
- Fasting blood sugar
- Resting electrocardiographic results
- Maximum heart rate
- Exercise-induced angina
- ST depression and slope
- Number of major vessels
- Thalassemia status
Analyzes voice measurements including:
- Fundamental frequency variations
- Jitter and shimmer measurements
- Noise-to-harmonics ratios
- Nonlinear dynamical complexity measures
- Signal processing features
├── application.py # Main Streamlit application
├── requirements.txt # Python dependencies
├── diabetes_model.pkl # Trained diabetes prediction model
├── heart_model.pkl # Trained heart disease prediction model
├── parkinsons_model.pkl # Trained Parkinson's disease prediction model
├── standardized_diabetes.pkl # Feature scaler for diabetes data
├── standardized_heart.pkl # Feature scaler for heart disease data
├── standardized_parkinsons.pkl # Feature scaler for Parkinson's data
- Python 3.7 or higher
- pip package manager
git clone <repository-url>
cd multiple-disease-prediction-systempip install -r requirements.txtstreamlit run application.pyOpen your web browser and navigate to:
http://localhost:8501
- Launch the Application: Run the Streamlit app using the command above
- Navigate: Use the sidebar menu to select the disease prediction you want to perform
- Input Data: Fill in the required medical parameters in the form
- Get Prediction: Click the "Predict" button to receive instant results
- Test with Samples: Use the provided sample inputs to test the application
Each prediction page includes comprehensive test cases:
- Healthy Profile: Low glucose, normal BMI, younger age
- Borderline Risk: Moderate values across parameters
- High Risk: Elevated glucose, high BMI, multiple risk factors
- Low Risk: Optimal blood pressure, normal cholesterol
- Moderate Risk: Some elevated parameters
- High Risk: Multiple cardiovascular risk factors
- Healthy Voice: Normal frequency variations and harmonics
- Borderline: Slight voice irregularities
- Parkinson's Indicators: Significant voice tremor patterns
- Algorithms: The system uses trained classification models
- Feature Scaling: All inputs are standardized using pre-fitted scalers for optimal model performance
- Model Format: Models are serialized using Python's pickle module for efficient loading
- Streamlit: Web application framework
- Streamlit-option-menu: Enhanced navigation menu
- Scikit-learn: Machine learning library for model operations
- NumPy: Numerical computing for data processing
- Pickle: Model serialization and deserialization
- Medical Advisory: This tool is for educational and screening purposes only
- Not a Substitute: Results should not replace professional medical diagnosis
- Consult Healthcare Providers: Always seek medical advice for health concerns
- Accuracy Limitations: Model predictions are based on training data and may not be 100% accurate
- New Disease Models: Add new pickle files and update the navigation menu
- Enhanced UI: Modify the Streamlit interface in
application.py - Model Updates: Replace existing pickle files with retrained models
The models were trained on standard medical datasets. To retrain:
- Prepare your dataset with appropriate features
- Train using scikit-learn or similar ML libraries
- Save models and scalers using pickle
- Update the application to load new models
- Add more disease prediction models
- Implement model confidence scores
- Add data visualization for risk factors
- Include historical prediction tracking
- Deploy to cloud platforms (Heroku, AWS, etc.)
- Add user authentication and data storage
- Implement API endpoints for integration
Contributions are welcome! Please feel free to submit a Pull Request. For major changes:
- Fork the repository
- Create a feature branch (
git checkout -b feature/AmazingFeature) - Commit your changes (
git commit -m 'Add some AmazingFeature') - Push to the branch (
git push origin feature/AmazingFeature) - Open a Pull Request
This project is licensed under the MIT License - see the LICENSE file for details.
Developed as a comprehensive health screening tool combining machine learning with user-friendly web interfaces for accessible medical risk assessment.
If you encounter any issues or have questions:
- Check the sample inputs provided in each prediction page
- Ensure all required fields are filled correctly
- Verify that dependencies are properly installed
- Open an issue in the GitHub repository for technical problems
🔥 Ready to start predicting? Launch the app and explore the power of AI in healthcare screening!