respiratory rate

[rpursley8]

Well, your data is pretty clean so the derivative part can be left out. Just delete that function and connect through. You will have to adjust the filter inputs fh and fl to give you a clean signal (fl is used to remove the DC drift and fh filters out the high frequency noise). For your data fl = 0.005 and fh=0. Adjust your threshold to around 0.2 and the width to around 10.

This should give you the indices for all of the peaks.

FYI, the number fh and fl relate to the sampling frequency, fs. When that isn't connected, fs=1. So if your sampling frequency is 100 Hz and you fed this number into fs, then fl=100*0.005=0.5 Hz and fh=100*0.02=2 Hz

[chilly_charly]

Has Randall said, your data are clean, and there is no reason to obtain a negative value. I suspect some mistake not related to the algorithm functionality.
One thing you could do is to have a look at the values returned by the peek detect vi. This vi returns an array of indices, but you use only the two first values. Modify slightly your vi (by the way try to arrange it a bit...), to be able to read all the peak indices, as in the attached jpg, and tell us what the indices are. Try to compare with the actual peak positions.

CC

[rpursley8]

For similar applications that I have had, I usually use the derivative function on the array of indices to get the difference between adjacent elements and those numbers can be converted from points to bpm.

[burcu]

Hi Charlie,

I've tried what you said. Enclosed you can find the indices,saved data and screenshots.

[burcu]

Here is the screen shot of front panel

[burcu]

Here is the screen shot of diagram

[burcu]

Here is the indices

[chilly_charly]

Sorry Burcu, but I don't understand what are Indices 1 and Indices 2. Indices should be a 1D-array, containing the same number of values than the number of peaks.

CC

[rpursley8]

Try modifying how you handle the indices as attached. The results should be correct.

Also, I like to use the other peak detect function because it gives an output for the amplitudes of each so I can plot this with my original data to verify I am capturing all of the peaks and only the peaks.

[burcu]

But indices are not 1-D array. I'm sending the .vi