Skip to content

Jaiden-Steele/ChestWiggleProject

BranchesTags

Repository files navigation

HFOV Chest Oscillation Monitor

Python 3.8+ Arduino

A real-time monitoring system for quantifying chest wall oscillations in pediatric patients undergoing High-Frequency Oscillatory Ventilation (HFOV)

📋 Overview

The HFOV Chest Oscillation Monitor is a senior design project addressing an unmet clinical need in pediatric and neonatal intensive care units (PICU/NICU). Currently, respiratory therapists assess the "chest wiggle factor" (CWF) visually, making it highly subjective and prone to error. Our solution provides objective, quantitative measurements of chest oscillations during HFOV therapy.

The Problem

  • Current Method: Visual assessment of chest oscillations by respiratory therapists
  • Limitations: Subjective, inconsistent between observers, difficult to document
  • Impact: Uncertainty in ventilation effectiveness, risk of improper treatment

Our Solution

A wearable accelerometer-based device that:

  • ✅ Provides real-time, quantitative measurements of chest oscillations
  • ✅ Non-invasive and safe for use with infants
  • ✅ Displays filtered HFOV signals (5-15 Hz) with noise reduction
  • ✅ Visualizes vibration magnitude and frequency in real-time

🎯 Key Features

  • Real-time Monitoring: Live visualization of chest oscillations at 100 Hz sampling rate
  • Advanced Signal Processing: Butterworth bandpass filter (5-15 Hz) isolates HFOV signals
  • Multi-axis Tracking: Independent X, Y, Z acceleration monitoring with LED indicators
  • Flexible Data Input: Supports both live Arduino data and synthetic testing data
  • Clinical Insights: Displays peak amplitude, RMS vibration, and frequency analysis
  • User-Friendly Interface: Pause/resume, reset controls, and clear visual feedback

🏥 Clinical Context

Intended Use: Monitoring chest wall oscillations in PICU/NICU patients (≤5 years old) undergoing HFOV

User: Respiratory therapists in pediatric intensive care settings

HFOV Parameters: Typical oscillation frequency of 5-15 Hz (device optimized for 8.5 Hz)

🛠️ System Architecture

┌─────────────────┐      USB Serial       ┌──────────────────┐
│  MPU6050 IMU    │ ───────────────────► │   Python GUI     │
│  (Arduino)      │    100 Hz @ 115200   │  (Matplotlib)    │
│                 │       baud            │                  │
│ • 3-axis accel  │                       │ • Real-time plot │
│ • ±2g range     │                       │ • Bandpass filter│
│ • 20Hz DLPF     │                       │ • Stats display  │
│ • Calibration   │                       │ • Visual alerts  │
└─────────────────┘                       └──────────────────┘

📦 Hardware Requirements

  • Microcontroller: Arduino Uno/Nano or compatible
  • Sensor: MPU6050 6-axis IMU (accelerometer + gyroscope)
  • Indicators: 3x LEDs (Red, Green, Blue) for axis-specific alerts
  • Connections:
    • MPU6050 → Arduino I2C (SDA, SCL)
    • LEDs → Digital pins 8, 9, 10 (with appropriate resistors)

Wiring Diagram

MPU6050          Arduino
────────         ───────
VCC    ──────►   5V
GND    ──────►   GND
SDA    ──────►   A4 (SDA)
SCL    ──────►   A5 (SCL)

LEDs             Arduino
────             ───────
Red LED   ──────► Pin 8  (X-axis deviation)
Green LED ──────► Pin 9  (Y-axis deviation)
Blue LED  ──────► Pin 10 (Z-axis deviation)

💻 Software Requirements

Python Dependencies

numpy>=1.21.0
matplotlib>=3.5.0
scipy>=1.7.0
pyserial>=3.5

Arduino Libraries

  • Wire (built-in)
  • MPU6050 by Electronic Cats

🚀 Installation

1. Clone the Repository

git clone https://github.com/yourusername/hfov-monitor.git
cd hfov-monitor

2. Install Python Dependencies

pip install numpy matplotlib scipy pyserial

3. Install Arduino Libraries

  1. Open Arduino IDE
  2. Go to Sketch → Include Library → Manage Libraries
  3. Search for "MPU6050" by Electronic Cats
  4. Click Install

4. Upload Arduino Firmware

  1. Connect your Arduino via USB
  2. Open arduino/hfov_sensor.ino in Arduino IDE
  3. Select your board and port from Tools menu
  4. Click Upload (→)

📊 Usage

Quick Start

  1. Hardware Setup:

    • Ensure MPU6050 is securely connected to Arduino
    • Upload the appropriate Arduino sketch
    • Place sensor on test surface or patient (when approved for clinical use)

  2. Run Monitor:

    python hfov_monitor.py

  3. Auto-Detection:

    • The software will automatically detect your Arduino
    • If detection fails, it will run with synthetic data for testing

  4. Manual Port Selection (if needed):

    # Windows
monitor = HFOVMonitor(port='COM3')

# Linux/Mac
monitor = HFOVMonitor(port='/dev/ttyUSB0')

Keyboard Controls

Key Action
SPACE Pause/Resume monitoring
R Reset buffers and restart
Q Quit application

Data Format

Arduino Output (CSV format):

ax,ay,az                    # 3 values
timestamp,ax,ay,az          # 4 values (optional timestamp)

Where ax, ay, az are acceleration values in g units (e.g., 0.05, 1.02, -0.15)

📈 Understanding the Display

The monitor shows three synchronized plots:

1. Raw Accelerometer Data

  • Shows unfiltered X, Y, Z acceleration
  • Includes breathing, heartbeat, and HFOV oscillations
  • Noise and baseline drift visible

2. Filtered HFOV Signal (5-15 Hz)

  • Bandpass filtered to isolate HFOV frequencies
  • Clear oscillation patterns
  • Removes breathing (~0.3 Hz) and baseline drift

3. Vibration Magnitude Comparison

  • Gray line: Raw unfiltered magnitude
  • Magenta line: Filtered HFOV magnitude (5-15 Hz)
  • Units: millig (1 mg = 0.001 g)

Status Panel

┌────────────────────────────────────────────────────────────────────┐
│  Mode: LIVE        │  Peak Amplitude: 45.2 mg  │  RMS Vibration: 32.1 mg  │
│  Status: RUNNING   │  Buffer: 150/500          │  Time: 12.5s             │
└────────────────────────────────────────────────────────────────────┘

🔧 Arduino Firmware Options

Version 1: LED Indicator Firmware

File: arduino/led_indicator.ino

Features:

  • Real-time LED feedback for each axis
  • Calibration routine on startup
  • Threshold-based alerts (0.20g default)
  • Serial debug output

LED Behavior:

  • 🔴 Red LED: X-axis deviation (forward/back)
  • 🟢 Green LED: Y-axis deviation (side-to-side)
  • 🔵 Blue LED: Z-axis deviation (up/down)

Version 2: Data Streaming Firmware

File: arduino/data_streaming.ino

Features:

  • Optimized for Python GUI integration
  • 100 Hz sampling rate
  • Digital low-pass filter (20 Hz)
  • Calibrated output in g units
  • Magnitude and vibration calculations

Use Case: Primary firmware for the monitoring GUI

🔬 Technical Specifications

Parameter Value
Sampling Rate 100 Hz
HFOV Frequency Range 5-15 Hz
Accelerometer Range ±2g
Digital Filter Butterworth 4th order bandpass
Data Buffer 500 samples (5 seconds)
Display Window 150 samples (1.5 seconds)
Update Rate 20 FPS

🎓 User Needs Addressed

Based on ethnographic observations and interviews with respiratory therapists:

  1. ✅ Accuracy across patient variability: Calibration system adapts to different body sizes
  2. ✅ Secure application: Lightweight, medical-grade adhesive mounting
  3. ✅ Sterilization compatibility: Device design considers clinical hygiene protocols
  4. ✅ System integration: Serial output compatible with existing monitoring systems
  5. ✅ Compact design: Small form factor, non-obstructive
  6. ✅ Rapid deployment: Quick calibration and setup (<2 minutes)
  7. ✅ Motion tolerance: Advanced filtering maintains signal quality during patient movement
  8. ✅ Continuous monitoring: Buffer system prevents data loss
  9. ✅ Long battery life: Low-power sensor with efficient sampling

🧪 Testing

Synthetic Data Mode

Test the GUI without hardware:

from hfov_monitor import HFOVMonitor

monitor = HFOVMonitor()  # Auto-falls back to synthetic if no Arduino found
monitor.run()

The synthetic signal includes:

  • HFOV oscillations (8.5 Hz)
  • Breathing (0.3 Hz)
  • Heartbeat (1.2 Hz)
  • Random noise

Calibration Testing

  1. Place sensor on flat, stable surface
  2. Run Arduino firmware
  3. Keep sensor motionless during calibration prompt
  4. Verify LEDs respond to gentle tilting

📝 Future Development

  • Wireless data transmission (Bluetooth/WiFi)
  • Integration with hospital EHR systems
  • Machine learning for anomaly detection
  • Multi-patient monitoring dashboard
  • FDA regulatory submission preparation
  • Clinical trial data collection

⚠️ Disclaimer

This device is a prototype for educational and research purposes only.

  • Not FDA approved for clinical use
  • Not intended for diagnostic or therapeutic decisions
  • Requires clinical validation and regulatory approval before patient use
  • Always follow hospital protocols and physician guidance

👥 Team

Senior Design Project - University of Pittsburgh

  • Connor Rees - Project Lead
  • Isabella Hsia - Financial Lead
  • Jaiden Steele
  • Isa Lisi

Clinical Consultant: Former Respiratory Therapist with 35+ years experience

🙏 Acknowledgments

  • Respiratory therapists at UPMC for clinical insights
  • University of Pittsburgh Department of Biomedical Engineering
  • Electronic Cats for MPU6050 Arduino library
  • Open-source Python community

📧 Contact

For questions, collaboration, or clinical feedback:

Made with ❤️ for improving pediatric critical care

Report Bug · Request Feature · Documentation

About

Prototyping HFOV chest wiggle simulator and accelerometer detection device

Resources

Readme
Activity

Stars

0 stars

Watchers

0 watching

Forks

0 forks
Report repository

Releases

No releases published

Packages

 
 
 

Contributors

Languages