vitaldsp 0.1.4

A comprehensive toolkit for Digital Signal Processing in healthcare applications.

Healthcare DSP Toolkit

A comprehensive toolkit for Digital Signal Processing (DSP) in healthcare applications

This repository contains a comprehensive toolkit for Digital Signal Processing (DSP) in healthcare applications. It includes traditional DSP methods as well as advanced machine learning (ML) and deep learning (DL) inspired techniques. The toolkit is designed to process a wide range of physiological signals, such as ECG, EEG, PPG, and respiratory signals, with applications in monitoring, anomaly detection, and signal quality assessment.

🌐 Web Application

Try the vital-DSP toolkit directly in your browser! We've deployed a web interface that allows you to explore the toolkit's capabilities without any local installation.

🔗 Launch Web Application

⚠️ Important Note: This web application is hosted on Render's free tier, which has some limitations:

• The application may take a few seconds to start up (cold start)
• It spins down after 15 minutes of inactivity to save resources
• Please use only the provided sample datasets for testing
• For production use or large datasets, consider running the toolkit locally

📁 Sample Data: Use the sample datasets available in the repository:

• ECG Data: sample_data/ECG/ - Contains sample electrocardiogram signals
• PPG Data: sample_data/PPG/ - Contains sample photoplethysmography signals
• These sample files are optimized for the web application's free tier limitations

Features

• Filtering: Traditional filters (e.g., moving average, Gaussian, Butterworth) and advanced ML-inspired filters.
• Transforms: Fourier Transform, DCT, Wavelet Transform, and various fusion methods.
• Time-Domain Analysis: Peak detection, envelope detection, ZCR, and advanced segmentation techniques.
• Advanced Methods: EMD, sparse signal processing, Bayesian optimization, and more.
• Neuro-Signal Processing: EEG band power analysis, ERP detection, cognitive load measurement. (To be implemented)
• Respiratory Analysis: Automated respiratory rate calculation, sleep apnea detection, and multi-sensor fusion.
• Signal Quality Assessment: SNR calculation, artifact detection/removal, and adaptive methods.
• Monitoring and Alert Systems: Real-time anomaly detection, multi-parameter monitoring, and alert correlation.

Installation

You can install vitalDSP in two different ways:

Option 1: Install via pip

If you want the simplest installation method, you can install the latest version of vitalDSP directly from PyPI using pip:

pip install vitalDSP

Option 2: Install from the GitHub Repository

For those who prefer to have the latest version, including any recent updates that may not yet be available on PyPI, you can clone the repository and install it manually. Step 1: Clone the Repository First, clone the vitalDSP repository from GitHub to your local machine:

git clone https://github.com/Oucru-Innovations/vital-DSP.gi

Step 2: Navigate to the Project Directory Navigate to the directory where the repository was cloned:

cd vital-DSP

Step 3: Install with setup.py You can now install vitalDSP using the setup.py script:

python setup.py install

This method ensures that you are using the most up-to-date codebase from the repository.

Applications in Healthcare

vitalDSP can be applied across various healthcare use cases:

• Remote Patient Monitoring: Analyze ECG and PPG signals for real-time insights into patient health.
• Stress and Anxiety Detection: Monitor heart rate variability to assess stress levels.
• Sleep Apnea Detection: Use respiratory signals to identify breathing irregularities during sleep.

Usage

Please read the instruction in the documentation for detailed usage examples and module descriptions.

Example Notebooks on Google Colab

• Synthetic Data & Peak Detection: Peak Detection, Synthetic Signal (ECG, PPG).
• Health Report Analysis: Generate Health Report from ECG and PPG data.
• Signal Quality Indices: Signal Quality Assurance, SNR, Artifact Detection.

Documentation

Comprehensive documentation for each module is available in the docs/ directory, covering usage examples, API references, and more.

Contributing

We welcome contributions! Please read the CONTRIBUTING file for guidelines on how to contribute to this project.

Community and Support

Join our community to share ideas, ask questions, and get support:

• GitHub Discussions: Join the conversation
• Report Issues: Found a bug? Open an issue

License

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