How to mark the peaks of a signal based on some threshold value?

Actually I am tring to mark the peaks of an abdominal electrocardiogram signal based on threshold value. If the signal value is greater than some threshold value, then that particular time instant will be denoted as peak. Can anyone me in sorting out this problem?

Answer 1

[Star Strider]

First, remove as much baseline drift and high-frequency noise as possible. That will make everything easier.

I suggest using findpeaks with the 'MinPeakProminence' parameter set to a reasonable value (for example 5 for an electrocardiogram). Alternatively, the islocalmax function with an appropriate value for 'MinProminence'.

Both functions have options to output the prominence values they detect, making it easier to detect the prominence values for the areas of interest and set the prominence threshold.

[BIPIN SAMUEL]

Thank you @Star Strider, but I also need the time indices of the part of the signal where the peak is found.

[Star Strider]

My pleasure!

Each function returns the indices of the peaks, either as a numeric vector (findpeaks) or a logical vector (islocalmax). For findpeaks, just use the signal alone as the argument (not both the signal and the time vector). So to get both the peaks values and the time values where ‘time’ is the time vector:

findpeaks:

[pks,locs] = findpeaks(y, 'MinPeakProminence',some_value);

peak_values = pks;

peak_times = time(locs);

islocalmax:

Lv = islocalmax(y, 'MinProminence',some_other_value)

peak_values = y(Lv);

peak_times = time(Lv)

.

[BIPIN SAMUEL]

@Star StriderThank You

[BIPIN SAMUEL]

@Star StriderThank You

[Star Strider]

My pleasure!

If my Answer helped you solve your problem, please Accept it!

.