ECG measurment (frequency of ECG)

[Photo1234]

TL;DR

ECG measurement

I was studying ECG measurements and come across the frequency of ECG. I couldn't find for sure the frequency of what this is? Is it the frequency of sampling(times of measurement) or is it something else?

 

[dlgoff]

Summary: ECG measurement

Is it the frequency of sampling(times of measurement) or is it something else?

[QUOTE=https://en.wikipedia.org/wiki/Neural_oscillation]Neural oscillations, or brainwaves, are rhythmic or repetitive patterns of neural activity in the central nervous system.[/QUOTE]

 

[Photo1234]

Yes but this is EEG, not ECG.(I am asking for ECG which measures heart, EEG measures electrical activity of the brain)

 

[marcusl]

Your question is too vague. Please show us what you were reading so we can answer.

 

[Photo1234]

Example:

 

[marcusl]

The first shows the power spectra (that is, plots of the energy content within each Hz of frequency span) for the ECG and for its primary components, together with those for a couple of the significant noise sources. The second plot is a kind of schematic showing the regions of frequency (not power) spectra that several physiological sources occupy, together with those for two noise sources that define the "noise floor."

 

[Photo1234]

The second plot is a kind of schematic showing the regions of frequency (not power) spectra that several physiological sources occupy, together with those for two noise sources that define the "noise floor."

Yes, that is what I was asking about. But what is a physiological source that of an ECG that has frequency? Heartbeats have a frequency, but that is not the frequency of ECG.

I look a bit more and it seems to me that the frequency is the frequency of ECG analyzed with Fourier transform. So is this true? And what is the physical(physiological) meaning of that frequency and of this analysis in this case?

 

[marcusl]

Yes, a spectrum results from taking a Fourier transform. The Q wave arises from atrial contraction, the R wave from ventricular contraction, and the S-T signals from ventricular relaxation (repolarization). The frequency components basically reflect the speeds with which these events occur. Trying to diagnose heart disease from the ECG spectrum has not proven particularly useful despite lots of research and attempt.

 

[Photo1234]

But which events? Atrial contraction and ventricular contraction happen with the frequency of the heartbeat. Maybe on cellular level?

 

[marcusl]

What do you know about bioelectricity (the generation of the ECG in particular), signals, Fourier series and transforms? What is your educational status? This would help in gauging how to answer.

 

[Photo1234]

I am undergraduate biologist, I think I know the basics of bioelectricity, I do not know much about signals and I know Fourier series very superficially, I just know some basic ideas, almost no concrete math.

 

[Guineafowl]

I may be misunderstanding your question, but...

A P-wave takes a certain time to rise and fall - the time period. Frequency is simply given as 1/T(ime period). So the P-wave contributes that frequency component to the spectrum shown in your first picture.

 

[berkeman]

You can see the waveforms from a typical 12-lead EKG (or ECG) below. The sampling bandwidth of your data acquisition system needs to be high enough so that you can see all of the features of the waveforms, and low enough to limit the noise that comes into the sampling and ADC system.

http://m3.wyanokecdn.com/924e8d3e4e2855d2e892602e7878c3a5.jpg

 

[marcusl]

Basically, the "speed" with which the waveform changes determines its frequency components. Imagine making up the waveform with bits of sine waves of differing frequencies--low frequencies for the long and rounded T wave, high frequencies for the short and spiky R wave. These correlate to what you see in the spectrum.

 

[Photo1234]

But ECG is technically is not spectrum(doesn't measure EM waves). When we have waves is easy to know what frequency means.
With Fourier analysis, we can write any function as a sum of sin waves, if I understand that correctly, so ECG can be as well. Is the frequency of these waves the frequency of ECG?

And do the waves have any physical meaning or this is just mathematical operation?

 

[berkeman]

Is the frequency of these waves the frequency of ECG?

Basically, yes. But it is more accurate to say that the bandwidth of the ECG processing and recording circuittry must be wide enough to accurately follow the ECG time domain waveforms.

And do the waves have any physical meaning or this is just mathematical operation?

Of course they do. What do you know about cardiac rhythms so far? What have you read about ECG/EKG rhythm analysis? Can you label an ECG waveform with the P-Q-R-S-T points, and describe what is happening at each point and in the intervals between them?

https://en.wikipedia.org/wiki/Electrocardiography

 

[Guineafowl]

But ECG is technically is not spectrum(doesn't measure EM waves).

You can plot the spectrum of sound waves, which are not EM waves either. When you look at a trace of an ECG, you are watching voltage vary with time; A vinyl record is a mini-plot of the sound pressure wave with time.

The spectrum plots are just looking at the wave in the frequency domain, rather than the time domain.

 

[Photo1234]

Yes of curse I know the rhythm and basic physiology behind it. I was not asking abaut physical meaning of classical ECG, but about the frequency "spectrum".

So following this logic without Fourier transform we could not "know" frequency of ECG?

 

[berkeman]

Yes of curse I know the rhythm and basic physiology behind it. I was not asking abaut physical meaning of classical ECG, but about the frequency "spectrum".

Well if you understand the physiology of the heart and its muscle and nerve system, you understand what the upper frequency limits are of what you can read on the ECG display (hence the "spectrum"). Ignoring artifacts, of course.

So following this logic without Fourier transform we could not "know" frequency of ECG?

You can test the bandwidth other ways, without the Fourier transform. For example, you can use an adjustable lowpass filter and bring the breakpoint down from kHz down through 100s of Hz and lower, and observe the effect on the displated ECG waveform. You can also use an adjustable bandpass filter to selectively let different parts of the ECG waveform spectrum through, and measure the peak voltages that make it through versus center frequency of the bandpass filter.

 

[Photo1234]

Well if you understand the physiology of the heart and its muscle and nerve system, you understand what the upper frequency limits are of what you can read on the ECG display (hence the "spectrum"). Ignoring artifacts, of course.

I do not know that. I have never read physiology literature and frequency in that regard.
And if this sine waves from ECG are only mathematical process how can we get rid of them with filters? Makes no sense to me.

 

[marcusl]

You have received some excellent and direct answers to your questions. I believe that your confusion arises from a lack of understanding of basic concepts like "waveform," "frequency," "spectrum," "Fourier analysis," etc. so I strongly suggest you read some introductory materials on these topics. Then we'll be better able to help you.