Removing heart rate noise from EMG data

Hi all,

Im doing some signal processing of EMG data we obtained during our experiments, but im not able to remove the heart noise. The emg we obtained looks like this (zoomed in):

is it possible to remove this heart rate noise with a function or a script? Maybe someone knows an app or something like that.

Thanks in advance for the help.

[Tala]

The definition of "noise" is vastly different in different signal processing applications. For common mechanical systems (eg. vibrations) we typically use a lowpass filter <10-20 hz. For solid-fuild interactions, we typically allow upto 200 hz. I suggest circle the parts you want to get rid of (your definition of noise) and attach your array. Also, have you seen this?

[Star Strider]

‘is it possible to remove this heart rate noise with a function or a script?’

Probably not with these data. Ideally, use a separate channel to record the EKG and subtract it later, or use a reference electrode, and subtract that signal from the desired signal in the differential amplifiers prior to the ADC.

[Adam Danz]

@nigel moos, it's fairly clear what you mean by noise in this example but the purpose of code performing this task is to solve this problem across many examples where the noise probably appears differently.

For this reason, you must explain what noise is, how it appears, and how it differs from the signal. You need a rule that holds true across all (or most) samples where there is noise.

One explanation is that you expect the pulse rate to be nearly uniform and the instances of noise to be sparse releative to the frequency of pulses. A program could then estimate the pulse rate and use that to look for variations between expected pulses or how the signal differs during a pulse.