Evaluation of a Continuous ECG Quality Indicator Based on the Autocorrelation Function

Jonathan Moeyersons1, Dries Testelmans2, Bertien Buyse2, Rik Willems3, Sabine Van Huffel1, Carolina Varon4
1ESAT-STADIUS, KU Leuven, 2Department of Pneumology, UZ Leuven, 3Department of Cardiology, UZ Leuven, 4ESAT-STADIUS, KU Leuven, & imec


Abstract

Background: Most electrocardiogram (ECG) signal quality assessment algorithms focus on a two- or multi-level classification. However, it could be argued that signal quality would more naturally occupy a continuum of quality values. Therefore, in previous work we created a continuous quality assessment algorithm based on the autocorrelation function (ACF). This paper evaluates this algorithm on a simulated dataset with five noise levels and known signal-to-noise ratios (SNR).

Methods: The quality assessment algorithm was evaluated on a large subset of the ECG signals of a polysomnographic dataset. An in-house quality estimation algorithm, with stringent thresholds, was applied to locate the cleanest epochs. Hereafter, calibrated amounts of two types of realistic ECG noise from the MIT-BIH Noise Stress Test Database (NSTDB) were added. Both Electrode Motion (EM) and Movement Artefacts (MA) were considered. For each clean ECG epoch a noisy epoch was randomly selected and a calibrated amount of this noise epoch was added to the clean epoch. The resulting quality values were compared by a Kruskal-Wallis test.

Results: Using only three features and a binary training set, we have shown significant inter-level quality decreases for both types of added noise (p<0.01). Despite this finding, also significant intra-level quality differences were observed, indicating a change in response according to the type of noise (p<0.01).

Conclusion: It has been stated that a continuous quality annotation is hard to validate since it is difficult to create a test dataset. In this paper we tried to circumvent this issue by evaluating a previously developed ECG signal quality indication tool on a simulated dataset with five noise levels and known SNR’s. Despite the simplicity of the algorithm, only three features were used, we have shown significant quality decreases per noise level for both types of added noise.