NeuroSyn is a home-based physiotherapy assistant that uses a surface EMG armband to:

• Calibrate muscle–signal baselines with a guided “Calibration Assistant”
• Monitor real-time muscle activation strength via a progress bar
• Classify physiotherapy exercises (e.g., elbow flexion, shoulder abduction, leg raises) with an AI model
• Deliver on-screen cues, countdowns, and corrective feedback in an intuitive GUI
• Log timestamped predictions and status messages in a CRT-style console

1. Calibration Assistant
   • Click “OK” to start, per-exercise countdown, auto-advance on completion
2. Real-Time Signal Processing
   • Z-score normalization & RMS feature extraction
   • Configurable confidence threshold for model predictions
3. Exercise Classification
   • TensorFlow/Keras model supporting multi-class recognition of rehab movements
   • Easily extendable via ICON_PATHS & custom icons in NSEmedia/gestures/
4. Patient-Centric UI
   • High-contrast, large text (ITC Garamond Condensed)
   • Gamified visual cues & progress tracking
   • Dot-Matrix font logging overlay
5. Edge-AI Ready
   • Prepared for on-device inference (TinyML / TensorFlow Lite)
   • Optimized for low-latency, low-power microcontrollers

git clone https://github.com/phenom64/NeuroSyn.git
cd NeuroSyn/model-v4
pip install -r requirements.txt

1. Set MYO_ADDRESS in constants.py to your armband's identifier (MAC Address)
2. [For developers/Forks] Place exercise icons under NSEmedia/gestures/ and update ICON_PATHS in constants.py

python NSE-interfaceFX.py

1. Calibration: Follow on-screen prompts on Calibration Assistant
2. Exercise Session: Perform movements; observe live feedback

This repository is a fork of the Artemis Project originally developed by the University of Manchester’s Robotics Society. As a key member of this society, I architected and implemented its cross-platform PyQt6 user interface for the Artemis Project. NeuroSyn extends my work with physiotherapy-focused calibration, classification, and feedback modules for my final-year dissertation in Product Design Engineering.