Issue in table indexing

I have a table with 5 columns and 100000 values (ECG and PPG signals). I need to find peaks of the signal starting from a value on the x axis (x_ECG_IV(i)) inside a range of 40 values. My idea is to use a for loop like this:

for i=1:length(peaks_ECG_IV)
    A=x_ECG_IV(i) - 20*Tc :Tc: x_ECG_IV(i) + 20*Tc;
    if x_ECG_IV(i) + 20*Tc < x_ECG_IV(length(peaks_ECG_IV)) && x_ECG_IV(i) - 20*Tc > x_ECG_IV(1)
    [y_i,x_i] = findpeaks(DatiECGPPG.ECGLeadIV,A);
    end

The problem is that A, array of 40(41 actually) values in which i need to find the peak for each x_ECG_IV(i), has not the same dimension of "DatiECGPPG.ECGLeadIV", which is the whole signal. How can i find the corresponding y value to the A interval?

Answer 1

[Star Strider]

It would help to have the data, and a clearer description of what you want to do.

The findpeaks function does not have to have its inputs ‘windowed’ by time limits in order to find the appropriate peaks. Just give it each column in turn, and provide it with appropriate name-value pair arguments. My favourite is ‘MinPeakProminence’ since it will find the larger peaks and ignore the smaller peaks, and will even work correctly with signals with varying baselines.

I would do something like this (assuming the first column is the time vector):

T1 = readtable('YourTable.txt')

p = [ ... ]; % Prominence Vector For Each Column

for k = 2:size(T1,2)

[pks,locs] = findpeaks(T1{:,k}, 'MinPeakProminence', p(k));

pksc{k-1,:} = pks;

locsc{k-1,:} = locs;

end

Then process them afterwards. The ‘locsc’ values will index into the time vector and all the other vectors for each signal (matrix column or table variable). It is likely straightforward to determine the corresponding intervals.

.

[Matteo Svaldi]

I'll add some context: the ECG signal has been filtered in order to find the peaks and peaks location in terms of time. The task is to find the peaks in the original signal starting from the filtered ones. The professor asked specifically to find them in a 40 range values range. Where should i specify this?

[Star Strider]

If you have the peak values and peak locations, filter the peaks using ‘logical indexing’ —

s = rand(40,1);
[pks,locs] = findpeaks(s)

pks = 12×1

0.8990 0.9961 0.9980 0.9159 0.5342 0.7040 0.7958 0.8126 0.6548 0.8179

locs = 12×1

4 8 10 13 15 21 23 25 27 29

peakmin = 0.5;
peakmax = 0.7;
Lv = (pks>=peakmin) & (pks<=peakmax);                       % Logical Vector

selected_pks =  pks(Lv)

selected_pks = 2×1

0.5342 0.6548

selected_locs = locs(Lv)

selected_locs = 2×1

15 27

Beyond that, I have no idea what you want to do.

.

[Matteo Svaldi]

Thank you now it's clear, i misunderstood the task

[Star Strider]

As always, my pleasure!