ECG signal artifact removing .

I have ECG signal has .mat extension I want to make good Baseline wander artifact removing by code, I know how to make it by fdatool , I just want it by codes to enhance it more ?

Answer 1

[Star Strider]

Try this:

Fs = 1000;                                                  % Sampling Frequency (Hz)
Fn = Fs/2;                                                  % Nyquist Frequency (Hz)
Wp = [1.1 100]/Fn;                                          % Passband Frequency (Normalised)
Ws = [0.1 101]/Fn;                                          % Stopband Frequency (Normalised)
Rp =   1;                                                   % Passband Ripple (dB)
Rs = 150;                                                   % Stopband Ripple (dB)
[n,Ws] = cheb2ord(Wp,Ws,Rp,Rs);                             % Filter Order
[z,p,k] = cheby2(n,Rs,Ws);                                  % Filter Design
[sosbp,gbp] = zp2sos(z,p,k);                                % Convert To Second-Order-Section For Stability

figure(3)
freqz(sosbp, 2^16, Fs)                                      % Filter Bode Plot

filtered_signal = filtfilt(sosbp, gbp, original_signal);    % Filter Signal

Tweak the filter characteristics (most notably ‘Fs’ to be appropriate to your signal) to get the result you want. The sampling frequency ‘Fs’ is ideally above 250 Hz to use this filter.

[@MSM]

thank you for your answer But I forget to give some points Fs=360 sampling /sec , Baseline wandering Required highpass filter to remove artifact , the workspace on matlab indicate that the file have a variable 1 X 10000 double

[Star Strider]

My pleasure.

This is a bandpass filter that will remove baseline wander and high-frequency noise. EKG signals are characteristically band-limited on the interval of 0 to 100 Hz even with arrhythmias. My filter design will do what you want.

Substitute:

Fs = 360;

in my code. The filter I designed will work.

[@MSM]

I appreciate your response thank you .

[Star Strider]

As always, my pleasure.

[@MSM]

I still have question , how I could implement this code on the variable appearing on workspace , then subtract the difference between original noise signal and filtered signal

thank you

[Star Strider]

Use the filtfilt function with the filter:

filtered_signal = filtfilt(sosbp, gbp, original_signal);    % Filter Signal

That is all you need to do. No further subtraction or other steps are necessary.

The variable in your workspace you use I refer to here as ‘original_signal’.

[@MSM]

thank you , but its give me error

the filter appearing but the signal not filtered appearing also

they are at same folder

Error in filterBW (line 12)
filtered_signal = filtfilt(sosbp, gbp, ***** );    % Filter Signal

[Star Strider]

It would help to know the complete text of the error message.

You cannot use ‘*****’ for your original (unfiltered) signal. You have to import it into your workspace (read it from the file), give it a variable name, and then supply that variable name to the filtfilt function to filter it.

See the documentation on Data Import and Analysis (link) and Supported File Formats for Import and Export (link) for a list of the functions you can use to import your data, depending on the file format it is currently in.