The paper presents an algorithm that helps to detect and replace artifacts from the IBI signal from which we can calculate HRV features.
Since HRV is a popular parameter to determine individual’s cardiovascular condition it is very important to analyse HRV in a way that also is clinically useful.
For HRV feature analysis IBI signal plays an important role. Study shows that a small of amount of artifact in the signal could produce erroneous HRV analysis in particular time domain HRV features are sensitive to artifacts as shown in table 3.
It shows that error is higher for time domain features for instance, for SDNN mean absolute error is 135.748 which are much higher than the other frequency domain features.