How to extract the time interval from areas of interest from a signal?

Hello! I have a blood pressure signal that looks like this:

https://scontent-frt3-1.xx.fbcdn.net/hphotos-xfp1/v/t35.0-12/12235753_1663006193941174_746177023_o.jpg?oh=b00b341f327554927e598e49b596f72e&oe=56414AC7

and I want to find the time interval for each descent part ( in my case 4 intervals). But I don't know how since it doesn't have any pattern. If anyone can, please help me. Thank you.

[Star Strider]

It actually has a pattern, but you do not have a time vector in your file. You might be able to recover the time vector if you know what the heart rate was.

Answer 1

[Image Analyst]

First you have to filter out the outliers. How are you doing that? Perhaps a median filter might work if they're not too close together. Then you can just invert the image and use findpeaks(). I can't do it because you didn't attach the actual data.

[Nicoleta]

Here is Alexandra's actual data that u need ('presiune.mat') . Let us know about how your code looks and how does the final result look because some of us don't understand how to implement the code in matlab even if they partially or totally understood your explanation in plain words. Thanks.

[Image Analyst]

Loading the file doesn't work - it says it's corrupt. Can you save the data in an xlsx workbook or a csv file?

[Nicoleta]

Hey!Hope it works now.Let me know about your further results. Thank you for your help.

[Image Analyst]

Something went wrong with your original picture you attached. Please fix it.

[Image Analyst]

Try this. Let me know how it goes:

signal = xlsread('presiune.xlsx');
plot(signal, 'b-');
[values, indexes] = findpeaks(-signal, 'MinPeakProminence', 25);
grid on;
hold on;
stem(indexes, 50*ones(1, length(indexes)));