COMPGROUPS.NET | Search | Post Question | Groups | Stream | About | Register

### determining number of heart beats in ECG

• Follow

```hi , I am using the MIT-BIH Arrhythmia Database and was asked to get 10 consecutive minutes, and i did and then three colums, first for time, second for MLII and third for V5 and ended up with these

x = txt(1:86401,1);
y = txt(1:86401,2);
z = txt(1:86401,3);
hold on;
plot(x,y)
plot(x,z)

but i am asked to determine number of heart beats and duration of the time interval between successive heart beats, and this is without using a filter, how would i do that since data number is so large and heartbeats are very small?

thanx so much
```
 0

```"gedaa " <ghaida8686@gmail.com> wrote in message <impj3p\$b0p\$1@fred.mathworks.com>...
> hi , I am using the MIT-BIH Arrhythmia Database and was asked to get 10 consecutive minutes, and i did and then three colums, first for time, second for MLII and third for V5 and ended up with these
>
> x = txt(1:86401,1);
> y = txt(1:86401,2);
> z = txt(1:86401,3);
> hold on;
> plot(x,y)
> plot(x,z)
>
> but i am asked to determine number of heart beats and duration of the time interval between successive heart beats, and this is without using a filter, how would i do that since data number is so large and heartbeats are very small?
>
> thanx so much

Have you tried using Fourier analysis?

Wayne
```
 0

```"gedaa " <ghaida8686@gmail.com> wrote in message <impj3p\$b0p\$1@fred.mathworks.com>...
> hi , I am using the MIT-BIH Arrhythmia Database and was asked to get 10 consecutive minutes, and i did and then three colums, first for time, second for MLII and third for V5 and ended up with these
>
> x = txt(1:86401,1);
> y = txt(1:86401,2);
> z = txt(1:86401,3);
> hold on;
> plot(x,y)
> plot(x,z)
>
> but i am asked to determine number of heart beats and duration of the time interval between successive heart beats, and this is without using a filter, how would i do that since data number is so large and heartbeats are very small?
>
> thanx so much

Hi,

you could try findpeaks. I suggest working on a 30 sec - 1 min portion first until you get a handle of the function's option and after that apply it to the whole signal. You should also take into consideration that an irregular / pseudo periodic signal (which one may expect in this case) could cause abnormal results, so it's up to you to make it reliable.

Hope it helps.
Regards,
Florin
```
 0

2 Replies
640 Views

Similiar Articles:

7/11/2012 1:53:56 PM