Home

🧼 hdsemg-select 🧼

HDsEMG data cleaning tool

Welcome to the documentation for hdsemg-select, a sophisticated graphical user interface (GUI) application designed for selecting and analyzing HDsEMG channels from .mat files. This tool helps identify and exclude faulty channels (e.g., due to electrode misplacement or corrosion) and automatically flag potential artifacts like ECG contamination, power line noise (50/60Hz), or general signal anomalies.

Key Features¶

✅ Support for multiple file formats (.mat, .otb+, .otb4)
🧠 Intelligent grid detection and configuration
🖼 Comprehensive visualization tools
⚡️ Advanced artifact detection
💾 Structured data export
🔍 Detailed signal analysis capabilities

Installation Guide: Step-by-step instructions for setting up hdsemg-select
Getting Started: Learn how to get started with the application. The documentation will guide you through the first steps of using the hdsemg-select application to inspect and clean your high-density surface EMG (HD-sEMG) data.
Developer Guide: Information for contributors and developers

Core Functionality¶

Signal Visualization¶

Grid-based electrode visualization
Time-domain and frequency spectrum analysis
Multi-channel overview with pagination
Reference signal overlay capabilities

Channel Management¶

Manual and automatic channel selection
Amplitude-based selection with configurable thresholds
Custom label management
Comprehensive artifact flagging system

Data Processing¶

Automatic artifact detection
- ECG contamination identification
- Power line noise detection (50/60Hz)
- General signal anomaly detection
Signal view options (MP, SD, DD)
Action potential propagation analysis

Data Export¶

Structured JSON export with channel metadata
Automated cleaned .mat file generation
Comprehensive channel labeling system

Requirements¶

Python 3.8+
See requirements.txt for detailed dependencies
Compatible with Linux and Windows 11

Contributing¶

Contributions are welcome! If you'd like to improve hdsemg-select, please take a look at our Contributing Guide for details on how to get started.

Visit our GitHub repository to get started.