how to find peak to peak amplitude of an ecg signal

how to find peak to peak amplitude of an ecg signal for 300 qrs complexes of MIT_BIH atrial fibrillation database

[Daniel M]

This may help

https://www.mathworks.com/matlabcentral/answers/348675-how-to-calculate-peak-to-peak-amplitude-from-a-waveform

[sai lakshmi]

how to do it for 300 QRS complexes in an ecg signal which is already filtered

[Adam Danz]

You'll need to supply a screen shot of your data so we can see what you're working with.

[sai lakshmi]

load('07910m.mat');

x1=(val(1,:));

Fs=250;

[qrs_amp_raw,qrs_i_raw,delay]=pan_tompkin(x1,Fs,1);

%signal power

for i=2:300

qrs_amp_raw(i)=qrs_amp_raw(i-1)+qrs_amp_raw(i);

SUM=qrs_amp_raw(i);

end

avg_qrs_amp_raw=SUM/300;

S=avg_qrs_amp_raw/8;

physionet MIT_BIH afib database has been used ..

[sai lakshmi]

This is the paper i am working on..data means what do you need..the coding or the graphs

[Image Analyst]

That's quite a complicated paper for us to read (and I didn't). I'm sure the algorithm is way more than the dozen lines of code in your script. I suggest you contact the authors for code.

By the way, I don't even know what peak-to-peak amplitude means. Going along the x-axis from one peak to the next means that you go along time, not amplitude. So do you mean peak-to-peak time? Or do you want the difference in amplitudes (heights along the y axis) from one peak to the next and don't really care about the time difference - you just care about the height difference.

[sai lakshmi]

about the height difference in qrs complex

[Daniel M]

This was just a cursory search

https://www.mathworks.com/help/signal/examples/peak-analysis.html

https://www.mathworks.com/help/wavelet/ug/r-wave-detection-in-the-ecg.html

[Adam Danz]

@sai lakshmi,

sorry if my request was vague. What I was asking for is an image of what your data looks like when visualized/plotted. Here's an unrelated example I pulled from the internet, below but even this image lacks critical information such as axis labels.

Remember that we are working in the dark so in order to maximize the help you get from volunteers here, you need to provide us with enough information to develop a mental image of your problem before we can think of a solution. A picture is worth 1000 words (plus the 300+ words already included in comments above) so if we can see what kind of peaks you're working with, your axis limits, the range of values within your data, the scale of your data, etc, we will be much better equipt to think of solutions.

This phrase "ecg signal for 300 qrs complexes of MIT_BIH atrial fibrillation database" tells us virtually nothing - even if we are familiar with ecg signals. Our best chances to help you would require a lot of time searching the internet to figure out what that database is - and why would we do that?

So, as a mini lesson, keep in mind the need to define a solid question that contains all the background information needed and requires less than 1 minute to read and understand.

[sai lakshmi]

sorry sir,

i have detected the R point in my waveform through pan tompkin algorithm implementation

i wanted to find Q and S points ,so that for each of the QRS complex i would compute peak to peak amplitude,this was my query.

i have linked my basepaper and database too along with my matlab pics during the running session,in the last graph i have got my R points

[Daniel M]

Did you look at the link I posted? It's a MATLAB example that literally shows you how to do this with code.

[sai lakshmi]

i ave already gone through it,as my knowledge in matlab is poor,i couldn't figure it out properly,only R point is detected by me

[Daniel M]

But it gives you the code! What part didn't you understand?

[Image Analyst]

Was the code the exact code from that exact paper that he wants to implement?

sai, if you can't code up the paper, then the Mathworks consulting group can do it for you.

[Daniel M]

It was never clear that this paper was a requirement. The example I suggested will provide him the locations of the QRS peaks. To me it appeared that he found a paper that did what he wanted, but their method was quite complicated, and I thought I would give him an alternative, simpler approach.

Sai, if you are specifically trying to reproduce the works of this paper, I can only agree with Image Analyst in regards to looking into a consulting option.