Smoothing ECG signals based on the Iterative Unbiased FIR algorithm with a hybrid horizon

Abstract

The electrocardiogram (ECG) is a non-invasive procedure that provides vital informa- tion about the heart. So, an analysis of ECG measurements to extract morphological characteristics is essential. The main challenge is to remove undesired components of the ECG signal without modifying its principal waves. The wave QRS is one of the most damaged when the ECG is processed with different techniques. Hence, it is required an algorithm to remove the ECG artifacts without morphologically altering the QRS wave. In this work, it is developed and implemented the Iterative Unbiased FIR algorithm (I-UFIR) with an automatic hybrid horizon (H- UFIR) aiming to avoid the alteration of the QRS wave. The proposed process shows an error reduction compared with the traditional I-UFIR. This statement is based on the mean square error metric (mse) and the ECG measurements of a free repository of ECG data.