How to calculate the heart rate from an ECG signal?

232 ビュー (過去 30 日間)
HelpAStudent
HelpAStudent 2023 年 6 月 15 日
コメント済み: Cris LaPierre 2023 年 6 月 15 日
Hi, I have for example this ecg signal from how I did the plot in time:
load 'sig.mat'
fs = 300;
t=linspace(0,length(sig)/fs,length(sig));
figure()
plot(t,sig/max(sig));
title('Signal in Time Domain');
xlabel('Time(s)');
ylabel('Amplitude');
And from the plot I roughly calculate in the x axis the time between one peak (R wave) to the other. That is circa 1.5 s. If the distance between two R peaks in an ECG is approximately 1.5 seconds, I calculate the heart rate based on that interval.
Converting 1.5 seconds to milliseconds:
1.5 seconds * 1000 = 1500 milliseconds
Now, we can calculate the heart rate:
Heart Rate = 60,000 / 1500 = 40 beats per minute
Therefore, with a distance between two R peaks of approximately 1.5 seconds, the heart rate would be around 40 beats per minute.
There is a more precice way to calculate the heart rate from an ECG signal?
Attention this signal is not filtered. Is better to filter first? And if so, how?
  1 件のコメント
Cris LaPierre
Cris LaPierre 2023 年 6 月 15 日
No need to convert to milliseconds
HeartRate = 60/1.5 % seconds/min * beats/sec = beats/min
HeartRate = 40

サインインしてコメントする。

回答 (1 件)

Cris LaPierre
Cris LaPierre 2023 年 6 月 15 日
  2 件のコメント
HelpAStudent
HelpAStudent 2023 年 6 月 15 日
編集済み: Cris LaPierre 2023 年 6 月 15 日
with findpeaks it not help me to find just the R peaks that I want. Since if I do the mean of the distance between this peaks I can't find a normal heart rate.
load sig.mat
% plotting
fs=300;
t = linspace(0,length(sig)/fs,length(sig));
f = linspace(- fs/2, fs/2, length(sig));
figure()
plot(t,sig)
title("signal")
xlabel("seconds")
ylabel("A.U.")
figure()
plot(f,fftshift(abs(fft(sig))))
title("signal spectrum")
xlabel("Hz")
ylabel("Amplitude")
% filtering
sig_noDC = sig - mean(sig);
freq1 = 49;
freq2 = 51;
w1 = freq1/(fs/2);
w2 = freq2/(fs/2);
w = [w1 w2];
[b,a] = butter(3, w, "stop");
filtered_sig = filtfilt(b,a,sig_noDC);
freqX = 0.5;
wX = freqX/(fs/2);
[b,a] = butter(3,wX,"high");
filtered_sig = filtfilt(b,a,filtered_sig);
figure()
plot(t,filtered_sig)
title("filtered signal")
xlabel("seconds")
ylabel("A.U.")
figure()
plot(f,fftshift(abs(fft(filtered_sig))))
title("filtered signal spectrum")
xlabel("Hz")
ylabel("Amplitude")
[pks,locs,w,p] = findpeaks(sig); %find R peak
meanw = mean(w) %w is the width of all the peaks
meanw = 4.7879
heartRate = mean(w)*60
heartRate = 287.2740
Cris LaPierre
Cris LaPierre 2023 年 6 月 15 日
You will need to spend some time finding the appropriate settings for your findpeaks function call. It also wouldn't hurt to filter your signal, though with the prominence of the R peaks, that's proabably not necessary.
Consider doing this interactive in a live script using the Find Local Extrema live task.

サインインしてコメントする。

カテゴリ

Help Center および File ExchangeDescriptive Statistics についてさらに検索

製品


リリース

R2021a

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by