Finding RRI fluctuations (ECG) in 1s windows

This is the code I am running.

%ECG data

load('Dog_1_interictal_segment_0001.mat')

sf = 399.0698; %sampling frequency (Hz), samples per second

T = 1/sf;

timeSeries = T:T:600.8122;

%filtering

%y filtered data

d = designfilt('lowpassfir', ...

'PassbandFrequency',0.15,'StopbandFrequency',0.2, ...

'PassbandRipple',1,'StopbandAttenuation',60, ...

'DesignMethod','equiripple');

y = filtfilt(d,interictal_segment_1.data(1,:));

%findpeak for y signal in row 1

[pks,locs] = findpeaks(y(1,:),'MinPeakDistance',0.3);

I have found the local max for my signal y(1,:). However i would like to find fluctuations in the signal for successive 1 second, non-overlapping windows. Supposed that peaks 1,2,3,4 are successive local max.

Such that:

fluctuation1=peak2-peak1

fluctuation2=peak3-peak2

fluctuations are counted for a total length of 1 second and data is kept in window.1. Repeat for another 1s and keep data in window.2.

I am new to Matlab so any help is appreciated. Thank you.

Answer 1

[Peng Li]

Your question isn't quite clear. If you want to get fluctuations of subsequent points in a time series as you described above, you can use diff function to do the first order difference, or simply using locs(2:end) - locs(1:end-1) (locs in the variable you defined in your program).

The thing that is confusing is that you are trying to get RRI within a 1-s window. That doesn't make a lot of sense as usually the heart beats about once, or a bit more than twice per second when people exercise. You are not able to get any fluctuations from a single heart beat.

[Shamus Sim]

Thanks for your feedback!

Its actually ECG for the brain. Thats why T of waves are much shorter.

These sound like good suggestions:

diff function to do the first order difference, or simply using locs(2:end) - locs(1:end-1)

However, could it also include these characteristics?

fluctuations are counted for a total length of 1 second and data is kept in window.1. Repeat for another 1s and keep data in window.2.

[Shamus Sim]

I mean EEG signal not ECG. Sorry