Calculating calibration ECG device parameters

Hello,

I am using MPS 450 multiparmater simulator to generate ECG signal captured by ADS1298. I am trying to evaluate the performance of the device. How to test the features of the following signals to calibrate the device:

• Square waveforms—test linearity, amplitude, and frequency response.
• Pulse waveforms—sweep speed, stylus damping, sensitivity, gain and drift
• Sine waveforms— ECG sensitivity, frequency response (high and low), and accuracy.
• Triangle waveforms— testing the linearity of an ECG unit.

How to calculate the calibration parameters

[Saurabh]

Hi @Elzbieta,

While researching on this I found a File Exchange with the name ECG simulation using MATLAB, and the aim of this file Exchange is to generate all possible forms of ECG signals with the parameters specified by the user.

https://www.mathworks.com/matlabcentral/fileexchange/10858-ecg-simulation-using-matlab

Or can check out this research paper that speak about this:

https://www.researchgate.net/publication/258833667_Study_and_analysis_of_ECG_signal_using_MATLAB_LABVIEW_as_effective_tools

Answer 1

[William Rose]

@Elzbieta,

As you know, the MPS450 allows the user to specify various output waveforms. See table below, from the MPS450 operator's manual.

You connect the MPS450 output to the ADS1298 input and collect the data for the waveform(s) of interest. You save the data as a text file - see ADS1298 instructions for details on how to do that. TI recommend colle3cting at least 10 cycles of data. So for a 2 Hz square wave, at least 5 seconds of sata, and for a 100 Hz sine wave, at least 0.1 seconds of data, althoughn more would be fine. I reocmmend samplng at 500 Hz or 1 kHz, since that is a good rate for sampling a real ECG.

Now comes the interesting part: evaluating the data in the files collected with the ADS1298. For the square wave, I would look for overshoot and ringing. For the triangle, I woudl look for symmetry. For the sine waves, I would use the built-in matlab functions thd(), sfdr(), sinad(), snr().

Good luck!