EKG-EMG Signal Processing with Wavelet Transform

Overview

This project focuses on processing ECG (Electrocardiogram) signals using a Symlet wavelet transform in Python and collecting raw ECG data from the Olimex EKG/EMG shield. The script preprocesses ECG signals by applying a Stationary Wavelet Transform (SWT), analyzing the variance of the signal and its decomposition levels. Additionally, a Node.js script is included to handle real-time data collection from the Olimex shield, outputting the data in millivolts.

Dataset Features

The dataset used in this project includes real-time ECG signals collected via the Olimex EKG/EMG Shield. The key features of this dataset are:

• ECG Signal: The primary signal collected over a 60-second window at a sampling rate of 256 Hz.
• Wavelet Decomposition: Symlet wavelet transformation of the signal, including approximation and detail coefficients.
• Variance Analysis: Computation of variance at each decomposition level to analyze the signal's frequency components.

Project Goals

The main objectives of this project are:

1. ECG Signal Processing: To preprocess and analyze ECG signals using wavelet transforms.
2. Real-time Data Acquisition: To interface with the Olimex EKG/EMG Shield for collecting raw ECG data.
3. Variance Computation: To study the variance across different wavelet decomposition levels for further analysis.
4. Data Visualization: To provide visual representation of the original ECG signal and its wavelet-transformed components.

Tools Used

• Python Libraries:
  • numpy: For numerical computations, including signal processing and variance calculation.
  • matplotlib: For plotting the ECG signal and its decomposition.
  • pywt: For performing wavelet transformations on the ECG data.
• Node.js Libraries:
  • serialport: For interfacing with the Olimex EKG/EMG Shield via serial communication.
  • fs: For handling file operations to save ECG data.
  • os: For interacting with the operating system to manage file encoding and output.

How to Use

Python ECG Signal Processing

1. Load the ECG Data: Load your ECG data from a file, for example, einthoven32.txt, and transpose it to separate the time and signal components.

   data = np.loadtxt('./einthoven32.txt').T
   tempo, ecg = data[0], data[1]

2. Perform Wavelet Transform: Apply the Symlet wavelet (sym4) using the pywt.swt() function.

   coeffs = pywt.swt(ecg, wavelet='sym4', level=3, trim_approx=True, norm=True)

3. Analyze and Plot the Results: Plot the original ECG signal, its decomposition, and variance analysis.

   plt.plot(ecg)
   plt.legend(['Original signal'])
   fig, axes = plt.subplots(len(coeffs))
   # Continue plotting the wavelet decomposition components

4. Run the Script: Use the Python script to load and process ECG data, visualize the decomposition, and analyze the variance.

Node.js Real-time Data Acquisition

1. Install Required Packages:

   • Install the necessary dependencies using npm:

     npm install serialport @serialport/parser-byte-length

2. Configure the Serial Port:

   • Set the serial port and baud rate for your platform:
     • For Ubuntu: Use /dev/ttyACM0.
     • For Windows: Use COM3 or another appropriate COM port.
     • Adjust the sampling rate and measurement time as needed:

     var port = "/dev/ttyACM0"; // for Ubuntu
     var baudrate = 57600;
     var samplingRate = 256;
     var measurementTime = 60; // in seconds

3. Node.js Script Explanation:

   • The Node.js script reads real-time data from the Olimex EKG/EMG Shield, processes the data, and saves it to a file (einthoven3.txt). The steps involved are:
     • Serial Communication:
       • A serial connection is opened using SerialPort.
       • Data is received in byte packages, parsed by the ByteLengthParser to handle 6-byte packets.
     • ECG Data Handling:
       • The incoming byte array is converted to a voltage value (in millivolts) using the Olimex shield's gain and voltage range.
     • File Saving:
       • The processed ECG data is saved as a text file where each row contains the timestamp and corresponding millivolt value.

4. Run the Node.js Script:

   • Execute the script to start receiving ECG data from the Olimex EKG/EMG Shield and saving it to a file:

     node ecg_acquisition.js

5. Key Functions:

   • byteArrayToLong(byteArray): Converts the incoming byte array to a long integer.
   • convertToMilliVolt(value): Converts raw ECG data to millivolts based on the Olimex shield's gain and voltage range.
   • writeToFile(data): Saves the processed ECG data to a file.
   • handleData(data): Processes incoming data, packages it, and saves it when the measurement time has been reached.

6. Example Output:

   • The output file will contain rows in the following format:

     0.0039 0.524
     0.0078 0.527
     0.0117 0.531
     ...
     

This allows you to capture real-time ECG data and save it for further processing. If you find any errors, improvements or suggestions, please contact us using the links below. Your contribution is greatly appreciated!

For More Information

For more information, codes, tutorials, and exciting projects, visit the links below:

• Email: alves_lucasoliveira@usp.br
• GitHub: cyblx
• LinkedIn: Cyblx